The escalating debate on AI’s capabilities warrants developing reliable metrics to assess machine “intelligence.” Recently, many anecdotal examples were used to suggest that newer Large Language Models (LLMs) like ChatGPT and GPT-4 exhibit Neural Theory-of-Mind (N-ToM); however, prior work reached conflicting conclusions regarding those abilities. We investigate the extent of LLMs’ N-ToM through an extensive evaluation of 6 tasks and find that while LLMs exhibit certain N-ToM abilities, this behavior is far from being robust. We further examine the factors impacting performance on N-ToM tasks and discover that LLMs struggle with adversarial examples, indicating reliance on shallow heuristics rather than robust ToM abilities. We caution against drawing conclusions from anecdotal examples, limited benchmark testing, and using human-designed psychological tests to evaluate models.
Despite serving as the foundation models for a wide range of NLP benchmarks, pre-trained language models have shown limited capabilities of acquiring implicit commonsense knowledge from self-supervision alone, compared to learning linguistic and factual knowledge that appear more explicitly in the surface patterns in text. In this work, we introduce commonsense knowledge transfer, a framework to transfer the commonsense knowledge stored in a neural commonsense knowledge model to a general-purpose pre-trained language model. It first exploits general texts to form queries for extracting commonsense knowledge from the neural commonsense knowledge model and then refines the language model with two self-supervised objectives: commonsense mask infilling and commonsense relation prediction, which align human language with the underlying commonsense knowledge. Empirical results show that our approach consistently improves the model’s performance on downstream tasks that require commonsense reasoning. Moreover, we find that the improvement is more significant in the few-shot setting. This suggests that our approach helps language models better transfer to downstream tasks without extensive supervision by injecting commonsense knowledge into their parameters.
Pre-trained Transformer models like T5 and BART have advanced the state of the art on a wide range of text generation tasks. Compressing these models into smaller ones has become critically important for practical use. Common neural network compression techniques such as knowledge distillation or quantization are limited to static compression where the compression ratio is fixed. In this paper, we introduce Modular Transformers, a modularized encoder-decoder framework for flexible sequence-to-sequence model compression. Modular Transformers trains modularized layers that have the same function of two or more consecutive layers in the original model via module replacing and knowledge distillation. After training, the modularized layers can be flexibly assembled into sequence-to-sequence models that meet different performance-efficiency trade-offs. Experimental results show that after a single training phase, by simply varying the assemble strategy, Modular Transformers can achieve flexible compression ratios from 1.1x to 6x with little to moderate relative performance drop.
We present NovaCOMET, an open commonsense knowledge model, that combines the best aspects of knowledge and general task models. Compared to previous knowledge models, NovaCOMET allows open-format relations enabling direct application to reasoning tasks; compared to general task models like Flan-T5, it explicitly centers knowledge, enabling superior performance for commonsense reasoning. NovaCOMET leverages the knowledge of opaque proprietary models to create an open knowledge pipeline. First, knowledge is symbolically distilled into NovATOMIC, a publicly-releaseddiscrete knowledge graph which can be audited, critiqued, and filtered. Next, we train NovaCOMET on NovATOMIC by fine-tuning an open-source pretrained model. NovaCOMET uses an open-format training objective, replacing the fixed relation sets of past knowledge models, enabling arbitrary structures within the data to serve as inputs or outputs. The resulting generation model, optionally augmented with human annotation, matches or exceeds comparable open task models like Flan-T5 on a range of commonsense generation tasks. NovaCOMET serves as a counterexample to the contemporary focus on instruction tuning only, demonstrating a distinct advantage to explicitly modeling commonsense knowledge as well.
While text style transfer has many applications across natural language processing, the core premise of transferring from a single source style is unrealistic in a real-world setting. In this work, we focus on arbitrary style transfer: rewriting a text from an arbitrary, unknown style to a target style. We propose STEER: Unified Style Transfer with Expert Reinforcement, a unified frame-work developed to overcome the challenge of limited parallel data for style transfer. STEER involves automatically generating a corpus of style-transfer pairs using a product of experts during decoding. The generated offline data is then used to pre-train an initial policy before switching to online, off-policy reinforcement learning for further improvements via fine-grained reward signals. STEER is unified and can transfer to multiple target styles from an arbitrary, unknown source style, making it particularly flexible and efficient. Experimental results on a challenging dataset with text from a diverse set of styles demonstrate state-of-the-art results compared to competitive baselines. Remarkably, STEER outperforms the 175B parameter instruction-tuned GPT-3 on overall style transfer quality, despite being 226 times smaller in size. We also show STEER is robust, maintaining its style transfer capabilities on out-of-domain data, and surpassing nearly all baselines across various styles. The success of our method highlights the potential of RL algorithms when augmented with controllable decoding to overcome the challenge of limited data supervision.
Large language models (LLMs) demonstrate an amazing proficiency and fluency in the use of language. Does that mean that they have also acquired insightful linguistic knowledge about the language, to an extent that they can serve as an “expert linguistic annotator’? In this paper, we examine the successes and limitations of the GPT-3, ChatGPT, and GPT-4 models, focusing on the Abstract Meaning Representation (AMR) parsing formalism (Banarescu et al., 2013), which provides rich graphical representations of sentence meaning structure while abstracting away from surface forms. We compare models’ analysis of this semantic structure across two settings: 1) direct production of AMR parses based on zero- and few-shot examples, and 2) indirect partial reconstruction of AMR via metalinguistic natural language queries (e.g., “Identify the primary event of this sentence, and the predicate corresponding to that event.”). Across these settings, we find that models can reliably reproduce the basic format of AMR, as well as some core event, argument, and modifier structure-however, model outputs are prone to frequent and major errors, and holistic analysis of parse acceptability shows that even with few-shot demonstrations, models have virtually 0% success in producing fully accurate parses. Eliciting responses in natural language produces similar patterns of errors. Overall, our findings indicate that these models out-of-the-box can accurately identify some core aspects of semantic structure, but there remain key limitations in their ability to support fully accurate semantic analyses or parses.
Moral or ethical judgments rely heavily on the specific contexts in which they occur. Understanding varying shades of defeasible contextualizations (i.e., additional information that strengthens or attenuates the moral acceptability of an action) is critical to accurately represent the subtlety and intricacy of grounded human moral judgment in real-life scenarios. We introduce defeasible moral reasoning: a task to provide grounded contexts that make an action more or less morally acceptable, along with commonsense rationales that justify the reasoning. To elicit high-quality task data, we take an iterative self-distillation approach that starts from a small amount of unstructured seed knowledge from GPT-3 and then alternates between (1) self-distillation from student models; (2) targeted filtering with a critic model trained by human judgment (to boost validity) and NLI (to boost diversity); (3) self-imitation learning (to amplify the desired data quality). This process yields a student model that produces defeasible contexts with improved validity, diversity, and defeasibility. From this model we distill a high-quality dataset, 𝛿-Rules-of-Thumb, of 1.2M entries of contextualizations and rationales for 115K defeasible moral actions rated highly by human annotators 85.9% to 99.8% of the time. Using 𝛿-RoT we obtain a final student model that wins over all intermediate student models by a notable margin.
Twitter bot detection is vital in combating misinformation and safeguarding the integrity of social media discourse. While malicious bots are becoming more and more sophisticated and personalized, standard bot detection approaches are still agnostic to social environments (henceforth, communities) the bots operate at. In this work, we introduce community-specific bot detection, estimating the percentage of bots given the context of a community. Our method—BotPercent—is an amalgamation of Twitter bot detection datasets and feature-, text-, and graph-based models, adjusted to a particular community on Twitter. We introduce an approach that performs confidence calibration across bot detection models, which addresses generalization issues in existing community-agnostic models targeting individual bots and leads to more accurate community-level bot estimations. Experiments demonstrate that BotPercent achieves state-of-the-art performance in community-level Twitter bot detection across both balanced and imbalanced class distribution settings, presenting a less biased estimator of Twitter bot populations within the communities we analyze. We then analyze bot rates in several Twitter groups, including users who engage with partisan news media, political communities in different countries, and more. Our results reveal that the presence of Twitter bots is not homogeneous, but exhibiting a spatial-temporal distribution with considerable heterogeneity that should be taken into account for content moderation and social media policy making. The implementation of BotPercent is available at https://github.com/TamSiuhin/BotPercent.
Ambiguity is an intrinsic feature of natural language. Managing ambiguity is a key part of human language understanding, allowing us to anticipate misunderstanding as communicators and revise our interpretations as listeners. As language models are increasingly employed as dialogue interfaces and writing aids, handling ambiguous language is critical to their success. We capture ambiguity in a sentence through its effect on entailment relations with another sentence, and collect AmbiEnt, a linguist-annotated benchmark of 1,645 examples with diverse kinds of ambiguity. We design a suite of tests based on AmbiEnt, presenting the first evaluation of pretrained LMs to recognize ambiguity and disentangle possible meanings. We find that the task remains extremely challenging, including for GPT-4, whose generated disambiguations are considered correct only 32% of the time in crowdworker evaluation, compared to 90% for disambiguations in our dataset. Finally, to illustrate the value of ambiguity-sensitive tools, we show that a multilabel NLI model can flag political claims in the wild that are misleading due to ambiguity. We encourage the field to rediscover the importance of ambiguity for NLP.
Commonsense norms are defeasible by context: reading books is usually great, but not when driving a car. While contexts can be explicitly described in language, in embodied scenarios, contexts are often provided visually. This type of visually grounded reasoning about defeasible commonsense norms is generally easy for humans, but (as we show) poses a challenge for machines, as it necessitates both visual understanding and reasoning about commonsense norms. We construct a new multimodal benchmark for studying commonsense norms: NormLens. NormLens consists of 10K human judgments accompanied by free-form explanations covering 2K multimodal situations, and serves as a probe to address two questions: (1) to what extent can models align with average human judgment? and (2) how well can models explain their predicted judgments? We find that state-of-the-art model judgments and explanations are not well-aligned with human annotation. Additionally, we present a simple yet effective approach to better align models with humans by distilling social commonsense knowledge from large language models. The data and code will be released.
Today’s language models can be remarkably intelligent yet still produce text that contains trivial commonsense errors. Therefore, we seek a retrospective verification approach that can reflect on the commonsense plausibility of the machine text, and introduce Vera, a general-purpose model that learns to estimate the commonsense plausibility of declarative statements. To support diverse commonsense domains, Vera is trained on ~7M commonsense statements that are automatically converted from 19 QA datasets and two commonsense knowledge bases, and using a combination of three training objectives. When applied to solving commonsense problems in the verification format, Vera substantially outperforms existing models that can be repurposed for commonsense verification, even including GPT-3.5/ChatGPT/GPT-4, and it further exhibits generalization capabilities to unseen tasks and provides well-calibrated outputs. We find that Vera excels at filtering machine-generated commonsense knowledge and is useful in detecting erroneous commonsense statements generated by models like ChatGPT in real-world settings.
While extreme-scale language models have demonstrated exceptional performance on a variety of language tasks, the degree of control over these language models through pure prompting can often be limited. Directly fine-tuning such language models can be effective for tailoring them, but it can be either extremely costly (e.g., GPT-3) or not even feasible for the broader community (e.g., GPT-4). We propose Inference-time Policy Adapters (IPA), which efficiently tailors a language model such as GPT-3 without fine-tuning it. IPA guides a large base model during decoding time through a lightweight policy adapter trained to optimize an arbitrary user objective with reinforcement learning. On five challenging text generation tasks, such as toxicity reduction and lexically constrained generation, IPA consistently brings significant improvements over off-the-shelf language models. It outperforms competitive baseline methods, sometimes even including expensive fine-tuning. In particular, tailoring GPT-2 with IPA can outperform GPT-3, while tailoring GPT-3 with IPA brings a major performance boost over GPT-3 (and sometimes even over GPT-4). Our promising results highlight the potential of IPA as a lightweight alternative to tailoring extreme-scale language models.
Extensive work has shown that the performance and interpretability of commonsense reasoning can be improved via knowledge-augmented reasoning methods, where the knowledge that underpins the reasoning process is explicitly verbalized and utilized. However, existing implementations, including “chain-of-thought” and its variants, fall short in capturing the *introspective* nature of knowledge required in commonsense reasoning, and in accounting for the mutual adaptation between the generation and utilization of knowledge. We propose a novel method to develop an introspective commonsense reasoner, **Crystal**. To tackle commonsense problems, it first introspects for knowledge statements related to the given question, and subsequently makes an informed prediction that is grounded in the previously introspected knowledge. The knowledge introspection and knowledge-grounded reasoning modes of the model are tuned via reinforcement learning to mutually adapt, where the reward derives from the feedback given by the model itself. Experiments show that Crystal significantly outperforms both the standard supervised finetuning and chain-of-thought distilled methods, and enhances the transparency of the commonsense reasoning process. Our work ultimately validates the feasibility and potential of reinforcing a neural model with self-feedback.
Data scarcity has been a long standing issue in the field of open-domain social dialogue. To quench this thirst, we present SODA: the first publicly available, million-scale high-quality social dialogue dataset. By contextualizing social commonsense knowledge from a knowledge graph, we are able to distill an exceptionally broad spectrum of social interactions from a large language model. Human evaluation shows that conversations in SODA are more consistent, specific, and (surprisingly) natural than those in prior human-authored datasets. Using SODA, we train COSMO: a generalizable conversation model that is significantly more natural and consistent on unseen datasets than best-performing conversation models (e.g., GODEL, BlenderBot-1, Koala, Vicuna). Experiments reveal COSMO is sometimes even preferred to the original human-written gold responses. Additionally, our results shed light on the distinction between knowledge-enriched conversations and natural social chitchats. We plan to make our data, model, and code public.
Theory of mind (ToM) evaluations currently focus on testing models using passive narratives that inherently lack interactivity. We introduce FANToM, a new benchmark designed to stress-test ToM within information-asymmetric conversational contexts via question answering. Our benchmark draws upon important theoretical requisites from psychology and necessary empirical considerations when evaluating large language models (LLMs). In particular, we formulate multiple types of questions that demand the same underlying reasoning to identify illusory or false sense of ToM capabilities in LLMs. We show that FANToM is challenging for state-of-the-art LLMs, which perform significantly worse than humans even with chain-of-thought reasoning or fine-tuning.
Generics express generalizations about the world (e.g., birds can fly) that are not universally true (e.g., newborn birds and penguins cannot fly). Commonsense knowledge bases, used extensively in NLP, encode some generic knowledge but rarely enumerate such exceptions and knowing when a generic statement holds or does not hold true is crucial for developing a comprehensive understanding of generics. We present a novel framework informed by linguistic theory to generate exemplars—specific cases when a generic holds true or false. We generate ~19k exemplars for ~650 generics and show that our framework outperforms a strong GPT-3 baseline by 12.8 precision points. Our analysis highlights the importance of linguistic theory-based controllability for generating exemplars, the insufficiency of knowledge bases as a source of exemplars, and the challenges exemplars pose for the task of natural language inference.
Large neural networks can now generate jokes, but do they really “understand” humor? We challenge AI models with three tasks derived from the New Yorker Cartoon Caption Contest: matching a joke to a cartoon, identifying a winning caption, and explaining why a winning caption is funny. These tasks encapsulate progressively more sophisticated aspects of “understanding” a cartoon; key elements are the complex, often surprising relationships between images and captions and the frequent inclusion of indirect and playful allusions to human experience and culture. We investigate both multimodal and language-only models: the former are challenged with the cartoon images directly, while the latter are given multifaceted descriptions of the visual scene to simulate human-level visual understanding. We find that both types of models struggle at all three tasks. For example, our best multimodal models fall 30 accuracy points behind human performance on the matching task, and, even when provided ground-truth visual scene descriptors, human-authored explanations are preferred head-to-head over the best machine-authored ones (few-shot GPT-4) in more than 2/3 of cases. We release models, code, leaderboard, and corpus, which includes newly-gathered annotations describing the image’s locations/entities, what’s unusual in the scene, and an explanation of the joke.
Generating free-text rationales is a promising step towards explainable NLP, yet evaluating such rationales remains a challenge. Existing metrics have mostly focused on measuring the association between the rationale and a given label. We argue that an ideal metric should focus on the new information uniquely provided in the rationale that is otherwise not provided in the input or the label. We investigate this research problem from an information-theoretic perspective using conditional V-information (Hewitt et al., 2021). More concretely, we propose a metric called REV (Rationale Evaluation with conditional V-information), to quantify the amount of new, label-relevant information in a rationale beyond the information already available in the input or the label. Experiments across four benchmarks with reasoning tasks, including chain-of-thought, demonstrate the effectiveness of REV in evaluating rationale-label pairs, compared to existing metrics. We further demonstrate REV is consistent with human judgments on rationale evaluations and provides more sensitive measurements of new information in free-text rationales. When used alongside traditional performance metrics, REV provides deeper insights into models’ reasoning and prediction processes.
Chain-of-thought prompting (e.g., “Let’s think step-by-ste”) primes large language models to verbalize rationalization for their predictions. While chain-of-thought can lead to dramatic performance gains, benefits appear to emerge only for sufficiently large models (beyond 50B parameters). We show that orders-of-magnitude smaller models (125M—1.3B parameters) can still benefit from chain-of-thought prompting. To achieve this, we introduce Symbolic Chain-of-Thought Distillation (SCoTD), a method to train a smaller student model on rationalizations sampled from a significantly larger teacher model. Experiments across several commonsense benchmarks show that: 1) SCoTD enhances the performance of the student model in both supervised and few-shot settings, and especially for challenge sets; 2) sampling many reasoning chains per instance from the teacher is paramount; and 3) after distillation, student chain-of-thoughts are judged by humans as comparable to the teacher, despite orders of magnitude fewer parameters. We test several hypotheses regarding what properties of chain-of-thought samples are important, e.g., diversity vs. teacher likelihood vs. open-endedness. We release our corpus of chain-of-thought samples and code.
The potential social harms that large language models pose, such as generating offensive content and reinforcing biases, are steeply rising. Existing works focus on coping with this concern while interacting with ill-intentioned users, such as those who explicitly make hate speech or elicit harmful responses. However, discussions on sensitive issues can become toxic even if the users are well-intentioned. For safer models in such scenarios, we present the Sensitive Questions and Acceptable Response (SQuARe) dataset, a large-scale Korean dataset of 49k sensitive questions with 42k acceptable and 46k non-acceptable responses. The dataset was constructed leveraging HyperCLOVA in a human-in-the-loop manner based on real news headlines. Experiments show that acceptable response generation significantly improves for HyperCLOVA and GPT-3, demonstrating the efficacy of this dataset.
Among the remarkable emergent capabilities of large language models (LMs) is free-text rationalization; beyond certain scale, large LMs are capable of generating seemingly useful rationalizations, which in turn, can dramatically enhance their performances on leaderboards. This phenomenon raises a question: can machine generated rationales also be useful for humans, especially when lay humans try to answer questions based on those machine rationales? We observe that human utility of existing rationales is far from satisfactory and expensive to estimate with human studies. Existing metrics like task performance of the LM generating the rationales or similarity between generated and gold rationales are not good indicators of their human utility. While we observe that certain properties of rationales like conciseness and novelty are correlated with their human utility, estimating them without human involvement is challenging. We show that, by estimating a rationale’s helpfulness in answering similar unseen instances, we can measure its human utility to a better extent. We also translate this finding into an automated score, Gen-U, that we propose, which can help improve LMs’ ability to generate rationales with better human utility, while maintaining most of its task performance. Lastly, we release all code and collected data with this project.
Commonsense capabilities of pre-trained language models dramatically improve with scale, leading many to believe that scale is the only winning recipe. But is it? Here, we investigate an alternative that a priori seems impossible: can smaller language models (e.g., GPT-2) win over models that are orders of magnitude larger and better (e.g., GPT-3), if powered with novel commonsense distillation algorithms?The key intellectual challenge is to design a learning algorithm that achieve a competitive level of commonsense acquisition, without relying on the benefits of scale. In particular, we study generative models of commonsense knowledge, focusing on the task of generating generics, statements of commonsense facts about everyday concepts, e.g., birds can fly. We introduce I2D2, a novel commonsense distillation framework that loosely follows the Symbolic Knowledge Distillation of West et al. but breaks the dependence on the extreme-scale teacher model with two innovations: (1) the novel adaptation of NeuroLogic Decoding to enhance the generation quality of the weak, off-the-shelf language models, and (2) self-imitation learning to iteratively learn from the model’s own enhanced commonsense acquisition capabilities. Empirical results suggest that scale is not the only way, as novel algorithms can be a promising alternative. Moreover, our study leads to a new corpus of generics, Gen-A-tomic, that is the largest and highest quality available to date.
We propose a novel task, G4C, to study teacher-student natural language interactions in a goal-driven and grounded environment. Dungeons and Dragons (D&D), a role-playing game, provides an ideal setting to investigate such interactions. Here, the Dungeon Master (DM), i.e., the teacher, guides the actions of several players—students, each with their own personas and abilities—to achieve shared goals grounded in a fantasy world. Our approach is to decompose and model these interactions into (1) the DM’s intent to guide players toward a given goal; (2) the DM’s guidance utterance to the players expressing this intent; and (3) a theory-of-mind (ToM) model that anticipates the players’ reaction to the guidance one turn into the future. We develop a novel reinforcement learning (RL) method for training a DM that generates guidance for players by rewarding utterances where the intent matches the ToM-anticipated player actions. Human and automated evaluations show that a DM trained to explicitly model intents and incorporate ToM of the players using RL generates better-quality guidance that is 3x more likely to fulfill the DM’s intent than a vanilla natural language generation (NLG) approach.
Context is everything, even in commonsense moral reasoning. Changing contexts can flip the moral judgment of an action; Lying to a friend is wrong in general, but may be morally acceptable if it is intended to protect their life. We present ClarifyDelphi, an interactive system that learns to ask clarification questions (e.g., why did you lie to your friend?) in order to elicit additional salient contexts of a social or moral situation. We posit that questions whose potential answers lead to diverging moral judgments are the most informative. Thus, we propose a reinforcement learning framework with a defeasibility reward that aims to maximize the divergence between moral judgments of hypothetical answers to a question. Human evaluation demonstrates that our system generates more relevant, informative and defeasible questions compared to competitive baselines. Our work is ultimately inspired by studies in cognitive science that have investigated the flexibility in moral cognition (i.e., the diverse contexts in which moral rules can be bent), and we hope that research in this direction can assist both cognitive and computational investigations of moral judgments.
Theory of Mind (ToM)—the ability to reason about the mental states of other people—is a key element of our social intelligence. Yet, despite their ever more impressive performance, large-scale neural language models still lack basic theory of mind capabilities out-of-the-box. We posit that simply scaling up models will not imbue them with theory of mind due to the inherently symbolic and implicit nature of the phenomenon, and instead investigate an alternative: can we design a decoding-time algorithm that enhances theory of mind of off-the-shelf neural language models without explicit supervision? We present SymbolicToM, a plug-and-play approach to reason about the belief states of multiple characters in reading comprehension tasks via explicit symbolic representation. More concretely, our approach tracks each entity’s beliefs, their estimation of other entities’ beliefs, and higher-order levels of reasoning, all through graphical representations, allowing for more precise and interpretable reasoning than previous approaches. Empirical results on the well-known ToMi benchmark (Le et al., 2019) demonstrate that SymbolicToM dramatically enhances off-the-shelf neural networks’ theory of mind in a zero-shot setting while showing robust out-of-distribution performance compared to supervised baselines. Our work also reveals spurious patterns in existing theory of mind benchmarks, emphasizing the importance of out-of-distribution evaluation and methods that do not overfit a particular dataset.
Despite recent advances in detecting fake news generated by neural models, their results are not readily applicable to effective detection of human-written disinformation. What limits the successful transfer between them is the sizable gap between machine-generated fake news and human-authored ones, including the notable differences in terms of style and underlying intent. With this in mind, we propose a novel framework for generating training examples that are informed by the known styles and strategies of human-authored propaganda. Specifically, we perform self-critical sequence training guided by natural language inference to ensure the validity of the generated articles, while also incorporating propaganda techniques, such as appeal to authority and loaded language. In particular, we create a new training dataset, PropaNews, with 2,256 examples, which we release for future use. Our experimental results show that fake news detectors trained on PropaNews are better at detecting human-written disinformation by 3.62–7.69% F1 score on two public datasets.
Dogwhistles are coded expressions that simultaneously convey one meaning to a broad audience and a second, often hateful or provocative, meaning to a narrow in-group; they are deployed to evade both political repercussions and algorithmic content moderation. For example, the word “cosmopolitan” in a sentence such as “we need to end the cosmopolitan experiment” can mean “worldly” to many but also secretly mean “Jewish” to a select few. We present the first large-scale computational investigation of dogwhistles. We develop a typology of dogwhistles, curate the largest-to-date glossary of over 300 dogwhistles with rich contextual information and examples, and analyze their usage in historical U.S. politicians’ speeches. We then assess whether a large language model (GPT-3) can identify dogwhistles and their meanings, and find that GPT-3’s performance varies widely across types of dogwhistles and targeted groups. Finally, we show that harmful content containing dogwhistles avoids toxicity detection, highlighting online risks presented by such coded language. This work sheds light on the theoretical and applied importance of dogwhistles in both NLP and computational social science, and provides resources to facilitate future research in modeling dogwhistles and mitigating their online harms.
Text detoxification has the potential to mitigate the harms of toxicity by rephrasing text to remove offensive meaning, but subtle toxicity remains challenging to tackle. We introduce MaRCo, a detoxification algorithm that combines controllable generation and text rewriting methods using a Product of Experts with autoencoder language models (LMs). MaRCo uses likelihoods under a non-toxic LM (expert) and a toxic LM (anti-expert) to find candidate words to mask and potentially replace. We evaluate our method on several subtle toxicity and microaggressions datasets, and show that it not only outperforms baselines on automatic metrics, but MaRCo’s rewrites are preferred 2.1 times more in human evaluation. Its applicability to instances of subtle toxicity is especially promising, demonstrating a path forward for addressing increasingly elusive online hate.
Figurative language is ubiquitous in English. Yet, the vast majority of NLP research focuses on literal language. Existing text representations by design rely on compositionality, while figurative language is often non- compositional. In this paper, we study the interpretation of two non-compositional figurative languages (idioms and similes). We collected datasets of fictional narratives containing a figurative expression along with crowd-sourced plausible and implausible continuations relying on the correct interpretation of the expression. We then trained models to choose or generate the plausible continuation. Our experiments show that models based solely on pre-trained language models perform substantially worse than humans on these tasks. We additionally propose knowledge-enhanced models, adopting human strategies for interpreting figurative language types: inferring meaning from the context and relying on the constituent words’ literal meanings. The knowledge-enhanced models improve the performance on both the discriminative and generative tasks, further bridging the gap from human performance.
What kinds of instructional prompts are easier to follow for Language Models (LMs)? We study this question by conducting extensive empirical analysis that shed light on important features of successful instructional prompts. Specifically, we study several classes of reframing techniques for manual reformulation of prompts into more effective ones. Some examples include decomposing a complex task instruction into multiple simpler tasks or itemizing instructions into sequential steps. Our experiments compare the zero-shot and few-shot performance of LMs prompted with reframed instructions on 12 NLP tasks across 6 categories. Compared with original instructions, our reframed instructions lead to significant improvements across LMs with different sizes. For example, the same reframed prompts boost few-shot performance of GPT3-series and GPT2-series by 12.5% and 6.7% respectively averaged over all tasks. Furthermore, reframed instructions reduce the number of examples required to prompt LMs in the few-shot setting. We hope these empirically-driven techniques will pave the way towards more effective future prompting algorithms.
Despite recent success, large neural models often generate factually incorrect text. Compounding this is the lack of a standard automatic evaluation for factuality–it cannot be meaningfully improved if it cannot be measured. Grounded generation promises a path to solving both of these problems: models draw on a reliable external document (grounding) for factual information, simplifying the challenge of factuality. Measuring factuality is also simplified–to factual consistency, testing whether the generation agrees with the grounding, rather than all facts. Yet, without a standard automatic metric for factual consistency, factually grounded generation remains an open problem. We study this problem for content transfer, in which generations extend a prompt, using information from factual grounding. Particularly, this domain allows us to introduce the notion of factual ablation for automatically measuring factual consistency: this captures the intuition that the model should be less likely to produce an output given a less relevant grounding document. In practice, we measure this by presenting a model with two grounding documents, and the model should prefer to use the more factually relevant one. We contribute two evaluation sets to measure this. Applying our new evaluation, we propose multiple novel methods improving over strong baselines.
While counterfactual data augmentation offers a promising step towards robust generalization in natural language processing, producing a set of counterfactuals that offer valuable inductive bias for models remains a challenge. Most existing approaches for producing counterfactuals, manual or automated, rely on small perturbations via minimal edits, resulting in simplistic changes. We introduce NeuroCounterfactuals, designed as loose counterfactuals, allowing for larger edits which result in naturalistic generations containing linguistic diversity, while still bearing similarity to the original document. Our novel generative approach bridges the benefits of constrained decoding, with those of language model adaptation for sentiment steering. Training data augmentation with our generations results in both in-domain and out-of-domain improvements for sentiment classification, outperforming even manually curated counterfactuals, under select settings. We further present detailed analyses to show the advantages of NeuroCounterfactuals over approaches involving simple, minimal edits.
While a substantial body of prior work has explored adversarial example generation for natural language understanding tasks, these examples are often unrealistic and diverge from the real-world data distributions. In this work, we introduce a two-stage adversarial example generation framework (NaturalAdversaries), for designing adversaries that are effective at fooling a given classifier and demonstrate natural-looking failure cases that could plausibly occur during in-the-wild deployment of the models. At the first stage a token attribution method is used to summarize a given classifier’s behavior as a function of the key tokens in the input. In the second stage a generative model is conditioned on the key tokens from the first stage. NaturalAdversaries is adaptable to both black-box and white-box adversarial attacks based on the level of access to the model parameters. Our results indicate these adversaries generalize across domains, and offer insights for future research on improving robustness of neural text classification models.
A recurring challenge of crowdsourcing NLP datasets at scale is that human writers often rely on repetitive patterns when crafting examples, leading to a lack of linguistic diversity. We introduce a novel approach for dataset creation based on worker and AI collaboration, which brings together the generative strength of language models and the evaluative strength of humans. Starting with an existing dataset, MultiNLI for natural language inference (NLI), our approach uses dataset cartography to automatically identify examples that demonstrate challenging reasoning patterns, and instructs GPT-3 to compose new examples with similar patterns. Machine generated examples are then automatically filtered, and finally revised and labeled by human crowdworkers. The resulting dataset, WANLI, consists of 107,885 NLI examples and presents unique empirical strengths over existing NLI datasets. Remarkably, training a model on WANLI improves performance on eight out-of-domain test sets we consider, including by 11% on HANS and 9% on Adversarial NLI, compared to training on the 4x larger MultiNLI. Moreover, it continues to be more effective than MultiNLI augmented with other NLI datasets. Our results demonstrate the promise of leveraging natural language generation techniques and re-imagining the role of humans in the dataset creation process.
Even to a simple and short news headline, readers react in a multitude of ways: cognitively (e.g. inferring the writer’s intent), emotionally (e.g. feeling distrust), and behaviorally (e.g. sharing the news with their friends). Such reactions are instantaneous and yet complex, as they rely on factors that go beyond interpreting factual content of news. We propose Misinfo Reaction Frames (MRF), a pragmatic formalism for modeling how readers might react to a news headline. In contrast to categorical schema, our free-text dimensions provide a more nuanced way of understanding intent beyond being benign or malicious. We also introduce a Misinfo Reaction Frames corpus, a crowdsourced dataset of reactions to over 25k news headlines focusing on global crises: the Covid-19 pandemic, climate change, and cancer. Empirical results confirm that it is indeed possible for neural models to predict the prominent patterns of readers’ reactions to previously unseen news headlines. Additionally, our user study shows that displaying machine-generated MRF implications alongside news headlines to readers can increase their trust in real news while decreasing their trust in misinformation. Our work demonstrates the feasibility and importance of pragmatic inferences on news headlines to help enhance AI-guided misinformation detection and mitigation.
It remains an open question whether incorporating external knowledge benefits commonsense reasoning while maintaining the flexibility of pretrained sequence models. To investigate this question, we develop generated knowledge prompting, which consists of generating knowledge from a language model, then providing the knowledge as additional input when answering a question. Our method does not require task-specific supervision for knowledge integration, or access to a structured knowledge base, yet it improves performance of large-scale, state-of-the-art models on four commonsense reasoning tasks, achieving state-of-the-art results on numerical commonsense (NumerSense), general commonsense (CommonsenseQA 2.0), and scientific commonsense (QASC) benchmarks. Generated knowledge prompting highlights large-scale language models as flexible sources of external knowledge for improving commonsense reasoning. Our code is available at github.com/liujch1998/GKP
Modern neural language models can produce remarkably fluent and grammatical text. So much, in fact, that recent work by Clark et al. (2021) has reported that conventional crowdsourcing can no longer reliably distinguish between machine-authored (GPT-3) and human-authored writing. As errors in machine generations become ever subtler and harder to spot, it poses a new challenge to the research community for robust machine text evaluation. We propose a new framework called Scarecrow for scrutinizing machine text via crowd annotation. To support the broad range of real machine errors that can be identified by laypeople, the ten error categories of Scarecrow—such as redundancy, commonsense errors, and incoherence—are identified through several rounds of crowd annotation experiments without a predefined ontology. We then use Scarecrow to collect over 41k error spans in human-written and machine-generated paragraphs of English language news text. We isolate factors for detailed analysis, including parameter count, training data, and various decoding-time configurations. Our approach successfully quantifies measurable gaps between human authored text and generations from models of several sizes, including fourteen configurations of GPT-3. In addition, our analysis unveils new insights, with detailed rationales provided by laypeople, e.g., that the commonsense capabilities have been improving with larger models while math capabilities have not, and that the choices of simple decoding hyperparameters can make remarkable differences on the perceived quality of machine text. We release our training material, annotation toolkit and dataset at https://yao-dou.github.io/scarecrow/.
Large language models are increasingly capable of generating fluent-appearing text with relatively little task-specific supervision. But can these models accurately explain classification decisions? We consider the task of generating free-text explanations using human-written examples in a few-shot manner. We find that (1) authoring higher quality prompts results in higher quality generations; and (2) surprisingly, in a head-to-head comparison, crowdworkers often prefer explanations generated by GPT-3 to crowdsourced explanations in existing datasets. Our human studies also show, however, that while models often produce factual, grammatical, and sufficient explanations, they have room to improve along axes such as providing novel information and supporting the label. We create a pipeline that combines GPT-3 with a supervised filter that incorporates binary acceptability judgments from humans in the loop. Despite the intrinsic subjectivity of acceptability judgments, we demonstrate that acceptability is partially correlated with various fine-grained attributes of explanations. Our approach is able to consistently filter GPT-3-generated explanations deemed acceptable by humans.
The dominant paradigm for neural text generation is left-to-right decoding from autoregressive language models. Constrained or controllable generation under complex lexical constraints, however, requires foresight to plan ahead feasible future paths. Drawing inspiration from the A* search algorithm, we propose NeuroLogic A*esque, a decoding algorithm that incorporates heuristic estimates of future cost. We develop lookahead heuristics that are efficient for large-scale language models, making our method a drop-in replacement for common techniques such as beam search and top-k sampling. To enable constrained generation, we build on NeuroLogic decoding (Lu et al., 2021), combining its flexibility in incorporating logical constraints with A*esque estimates of future constraint satisfaction. Our approach outperforms competitive baselines on five generation tasks, and achieves new state-of-the-art performance on table-to-text generation, constrained machine translation, and keyword-constrained generation. The improvements are particularly notable on tasks that require complex constraint satisfaction or in few-shot or zero-shot settings. NeuroLogic A*esque illustrates the power of decoding for improving and enabling new capabilities of large-scale language models.
We establish THumB, a rubric-based human evaluation protocol for image captioning models. Our scoring rubrics and their definitions are carefully developed based on machine- and human-generated captions on the MSCOCO dataset. Each caption is evaluated along two main dimensions in a tradeoff (precision and recall) as well as other aspects that measure the text quality (fluency, conciseness, and inclusive language). Our evaluations demonstrate several critical problems of the current evaluation practice. Human-generated captions show substantially higher quality than machine-generated ones, especially in coverage of salient information (i.e., recall), while most automatic metrics say the opposite. Our rubric-based results reveal that CLIPScore, a recent metric that uses image features, better correlates with human judgments than conventional text-only metrics because it is more sensitive to recall. We hope that this work will promote a more transparent evaluation protocol for image captioning and its automatic metrics.
Natural language processing researchers have identified limitations of evaluation methodology for generation tasks, with new questions raised about the validity of automatic metrics and of crowdworker judgments. Meanwhile, efforts to improve generation models tend to depend on simple n-gram overlap metrics (e.g., BLEU, ROUGE). We argue that new advances on models and metrics should each more directly benefit and inform the other. We therefore propose a generalization of leaderboards, bidimensional leaderboards (Billboards), that simultaneously tracks progress in language generation models and metrics for their evaluation. Unlike conventional unidimensional leaderboards that sort submitted systems by predetermined metrics, a Billboard accepts both generators and evaluation metrics as competing entries. A Billboard automatically creates an ensemble metric that selects and linearly combines a few metrics based on a global analysis across generators. Further, metrics are ranked based on their correlation with human judgments. We release four Billboards for machine translation, summarization, and image captioning. We demonstrate that a linear ensemble of a few diverse metrics sometimes substantially outperforms existing metrics in isolation. Our mixed-effects model analysis shows that most automatic metrics, especially the reference-based ones, overrate machine over human generation, demonstrating the importance of updating metrics as generation models become stronger (and perhaps more similar to humans) in the future.
Recent video-text models can retrieve relevant videos based on text with a high accuracy, but to what extent do they comprehend the semantics of the text? Can they discriminate between similar entities and actions? To answer this, we propose an evaluation framework that probes video-text models with hard negatives. We automatically build contrast sets, where true textual descriptions are manipulated in ways that change their semantics while maintaining plausibility. Specifically, we leverage a pre-trained language model and a set of heuristics to create verb and person entity focused contrast sets. We apply these in the multiple choice video to-text classification setting. We test the robustness of recent methods on the proposed automatic contrast sets, and compare them to additionally collected human-generated counterparts, to assess their effectiveness. We see that model performance suffers across all methods, erasing the gap between recent CLIP-based methods vs. the earlier methods.
Fine-tuning continuous prompts for target tasks has recently emerged as a compact alternative to full model fine-tuning. Motivated by these promising results, we investigate the feasibility of extracting a discrete (textual) interpretation of continuous prompts that is faithful to the problem they solve. In practice, we observe a “wayward” behavior between the task solved by continuous prompts and their nearest neighbor discrete projections: We can find continuous prompts that solve a task while being projected to an arbitrary text (e.g., definition of a different or even a contradictory task), while being within a very small (2%) margin of the best continuous prompt of the same size for the task. We provide intuitions behind this odd and surprising behavior, as well as extensive empirical analyses quantifying the effect of various parameters. For instance, for larger model sizes we observe higher waywardness, i.e, we can find prompts that more closely map to any arbitrary text with a smaller drop in accuracy. These findings have important implications relating to the difficulty of faithfully interpreting continuous prompts and their generalization across models and tasks, providing guidance for future progress in prompting language models.
Machines that can represent and describe environmental soundscapes have practical potential, e.g., for audio tagging and captioning. Prevailing learning paradigms of audio-text connections have been relying on parallel audio-text data, which is, however, scarcely available on the web. We propose VIP-ANT that induces Audio-Text alignment without using any parallel audio-text data. Our key idea is to share the image modality between bi-modal image-text representations and bi-modal image-audio representations; the image modality functions as a pivot and connects audio and text in a tri-modal embedding space implicitly. In a difficult zero-shot setting with no paired audio-text data, our model demonstrates state-of-the-art zero-shot performance on the ESC50 and US8K audio classification tasks, and even surpasses the supervised state of the art for Clotho caption retrieval (with audio queries) by 2.2% R@1. We further investigate cases of minimal audio-text supervision, finding that, e.g., just a few hundred supervised audio-text pairs increase the zero-shot audio classification accuracy by 8% on US8K. However, to match human parity on some zero-shot tasks, our empirical scaling experiments suggest that we would need about 221 ≈ 2M supervised audio-caption pairs. Our work opens up new avenues for learning audio-text connections with little to no parallel audio-text data.
The common practice for training commonsense models has gone from–human–to–corpus–to–machine: humans author commonsense knowledge graphs in order to train commonsense models. In this work, we investigate an alternative, from–machine–to–corpus–to–machine: general language models author these commonsense knowledge graphs to train commonsense models. Our study leads to a new framework, Symbolic Knowledge Distillation. As with prior art in Knowledge Distillation (Hinton et al. 2015), our approach uses larger models to teach smaller models. A key difference is that we distill knowledge symbolically–as text–in addition to the neural model. We distill only one aspect–the commonsense of a general language model teacher, allowing the student to be a different type, a commonsense model. Altogether, we show that careful prompt engineering and a separately trained critic model allow us to selectively distill high-quality causal commonsense from GPT-3, a general language model. Empirical results demonstrate that, for the first time, a human-authored commonsense knowledge graph is surpassed by our automatically distilled variant in all three criteria: quantity, quality, and diversity. In addition, it results in a neural commonsense model that surpasses the teacher model’s commonsense capabilities despite its 100x smaller size. We apply this to the ATOMIC resource, and will share our new symbolic knowledge graph and commonsense models.
The perceived toxicity of language can vary based on someone’s identity and beliefs, but this variation is often ignored when collecting toxic language datasets, resulting in dataset and model biases. We seek to understand the *who*, *why*, and *what* behind biases in toxicity annotations. In two online studies with demographically and politically diverse participants, we investigate the effect of annotator identities (*who*) and beliefs (*why*), drawing from social psychology research about hate speech, free speech, racist beliefs, political leaning, and more. We disentangle *what* is annotated as toxic by considering posts with three characteristics: anti-Black language, African American English (AAE) dialect, and vulgarity. Our results show strong associations between annotator identity and beliefs and their ratings of toxicity. Notably, more conservative annotators and those who scored highly on our scale for racist beliefs were less likely to rate anti-Black language as toxic, but more likely to rate AAE as toxic. We additionally present a case study illustrating how a popular toxicity detection system’s ratings inherently reflect only specific beliefs and perspectives. Our findings call for contextualizing toxicity labels in social variables, which raises immense implications for toxic language annotation and detection.
We focus on creating agents that act in alignment with socially beneficial norms and values in interactive narratives or text-based games—environments wherein an agent perceives and interacts with a world through natural language. Such interactive agents are often trained via reinforcement learning to optimize task performance, even when such rewards may lead to agent behaviors that violate societal norms—causing harm either to the agent itself or other entities in the environment. Social value alignment refers to creating agents whose behaviors conform to expected moral and social norms for a given context and group of people—in our case, it means agents that behave in a manner that is less harmful and more beneficial for themselves and others. We build on the Jiminy Cricket benchmark (Hendrycks et al. 2021), a set of 25 annotated interactive narratives containing thousands of morally salient scenarios covering everything from theft and bodily harm to altruism. We introduce the GALAD (Game-value ALignment through Action Distillation) agent that uses the social commonsense knowledge present in specially trained language models to contextually restrict its action space to only those actions that are aligned with socially beneficial values. An experimental study shows that the GALAD agent makes decisions efficiently enough to improve state-of-the-art task performance by 4% while reducing the frequency of socially harmful behaviors by 25% compared to strong contemporary value alignment approaches.
Pre-trained language models (LMs) struggle with consistent reasoning; recently, prompting LMs to generate explanations that self-guide the inference has emerged as a promising direction to amend this. However, these approaches are fundamentally bounded by the correctness of explanations, which themselves are often noisy and inconsistent. In this work, we develop Maieutic Prompting, which aims to infer a correct answer to a question even from the unreliable generations of LM. Maieutic Prompting induces a tree of explanations abductively (e.g. X is true, because ...) and recursively, then frames the inference as a satisfiability problem over these explanations and their logical relations. We test Maieutic Prompting for true/false QA on three challenging benchmarks that require complex commonsense reasoning. Maieutic Prompting achieves up to 20% better accuracy than state-of-the-art prompting methods, and as a fully unsupervised approach, performs competitively with supervised models. We also show that Maieutic Prompting improves robustness in inference while providing interpretable rationales.
Social intelligence and Theory of Mind (TOM), i.e., the ability to reason about the different mental states, intents, and reactions of all people involved, allows humans to effectively navigate and understand everyday social interactions. As NLP systems are used in increasingly complex social situations, their ability to grasp social dynamics becomes crucial. In this work, we examine the open question of social intelligence and Theory of Mind in modern NLP systems from an empirical and theorybased perspective. We show that one of today’s largest language models (GPT-3; Brown et al., 2020) lacks this kind of social intelligence out-of-the box, using two tasks: SocialIQa (Sap et al., 2019), which measure models’ ability to understand intents and reactions of participants of social interactions, and ToMi (Le, Boureau, and Nickel, 2019), which measures whether models can infer mental states and realities of participants of situations. Our results show that models struggle substantially at these Theory of Mind tasks, with well-below-human accuracies of 55% and 60% on SocialIQa and ToMi, respectively. To conclude, we draw on theories from pragmatics to contextualize this shortcoming of large language models, by examining the limitations stemming from their data, neural architecture, and training paradigms. Challenging the prevalent narrative that only scale is needed, we posit that person-centric NLP approaches might be more effective towards neural Theory of Mind.
Most existing dialogue systems fail to respond properly to potentially unsafe user utterances by either ignoring or passively agreeing with them. To address this issue, we introduce ProsocialDialog, the first large-scale multi-turn dialogue dataset to teach conversational agents to respond to problematic content following social norms. Covering diverse unethical, problematic, biased, and toxic situations, ProsocialDialog contains responses that encourage prosocial behavior, grounded in commonsense social rules (i.e., rules-of-thumb, RoTs). Created via a human-AI collaborative framework, ProsocialDialog consists of 58K dialogues, with 331K utterances, 160K unique RoTs, and 497K dialogue safety labels accompanied by free-form rationales. With this dataset, we introduce a dialogue safety detection module, Canary, capable of generating RoTs given conversational context, and a socially-informed dialogue agent, Prost. Empirical results show that Prost generates more socially acceptable dialogues compared to other state-of-the-art language and dialogue models in both in-domain and out-of-domain settings. Additionally, Canary effectively guides conversational agents and off-the-shelf language models to generate significantly more prosocial responses. Our work highlights the promise and importance of creating and steering conversational AI to be socially responsible.
Many language generation models are now available for a wide range of generation tasks, including machine translation and summarization. Combining such diverse models may lead to further progress, but ensembling generation models is challenging during inference: conventional ensembling methods (e.g., shallow fusion) require that the models share vocabulary/tokenization schemes. We introduce Twist decoding, a simple and general text generation algorithm that benefits from diverse models at inference time. Our method does not assume the vocabulary, tokenization or even generation order is shared. Our extensive evaluations on machine translation and scientific paper summarization demonstrate that Twist decoding substantially outperforms each model decoded in isolation over various scenarios, including cases where domain-specific and general-purpose models are both available. Twist decoding also consistently outperforms the popular reranking heuristic where output candidates from one model are rescored by another. We hope that our work will encourage researchers and practitioners to examine generation models collectively, not just independently, and to seek out models with complementary strengths to the currently available models.
Knowledge underpins reasoning. Recent research demonstrates that when relevant knowledge is provided as additional context to commonsense question answering (QA), it can substantially enhance the performance even on top of state-of-the-art. The fundamental challenge is where and how to find such knowledge that is high quality and on point with respect to the question; knowledge retrieved from knowledge bases are incomplete and knowledge generated from language models are inconsistent. We present Rainier, or Reinforced Knowledge Introspector, that learns to generate contextually relevant knowledge in response to given questions. Our approach starts by imitating knowledge generated by GPT-3, then learns to generate its own knowledge via reinforcement learning where rewards are shaped based on the increased performance on the resulting question answering. Rainier demonstrates substantial and consistent performance gains when tested over 9 different commonsense benchmarks: including 5 datasets that are seen during model training, as well as 4 datasets that are kept unseen. Our work is the first to report that knowledge generated by models that are orders of magnitude smaller than GPT-3, even without direct supervision on the knowledge itself, can exceed the quality of commonsense knowledge elicited from GPT-3.
We present Referee, a novel framework for sentence summarization that can be trained reference-free (i.e., requiring no gold summaries for supervision), while allowing direct control for compression ratio. Our work is the first to demonstrate that reference-free, controlled sentence summarization is feasible via the conceptual framework of Symbolic Knowledge Distillation (West et al., 2022), where latent knowledge in pre-trained language models is distilled via explicit examples sampled from the teacher models, further purified with three types of filters: length, fidelity, and Information Bottleneck. Moreover, we uniquely propose iterative distillation of knowledge, where student models from the previous iteration of distillation serve as teacher models in the next iteration. Starting off from a relatively modest set of GPT3-generated summaries, we demonstrate how iterative knowledge distillation can lead to considerably smaller, but better summarizers with sharper controllability. A useful by-product of this iterative distillation process is a high-quality dataset of sentence-summary pairs with varying degrees of compression ratios. Empirical results demonstrate that the final student models vastly outperform the much larger GPT3-Instruct model in terms of the controllability of compression ratios, without compromising the quality of resulting summarization.
While often assumed a gold standard, effective human evaluation of text generation remains an important, open area for research. We revisit this problem with a focus on producing consistent evaluations that are reproducible—over time and across different populations. We study this goal in different stages of the human evaluation pipeline. In particular, we consider design choices for the annotation interface used to elicit human judgments and their impact on reproducibility. Furthermore, we develop an automated mechanism for maintaining annotator quality via a probabilistic model that detects and excludes noisy annotators. Putting these lessons together, we introduce GENIE: a system for running standardized human evaluations across different generation tasks. We instantiate GENIE with datasets representing four core challenges in text generation: machine translation, summarization, commonsense reasoning, and machine comprehension. For each task, GENIE offers a leaderboard that automatically crowdsources annotations for submissions, evaluating them along axes such as correctness, conciseness, and fluency. We have made the GENIE leaderboards publicly available, and have already ranked 50 submissions from 10 different research groups. We hope GENIE encourages further progress toward effective, standardized evaluations for text generation.
We introduce a general framework for abstractive summarization with factual consistency and distinct modeling of the narrative flow in an output summary. Our work addresses current limitations of models for abstractive summarization that often hallucinate information or generate summaries with coherence issues. To generate abstractive summaries with factual consistency and narrative flow, we propose Cooperative Generator-Discriminator Networks (Co-opNet), a novel transformer-based framework where the generator works with a discriminator architecture to compose coherent long-form summaries. We explore four different discriminator objectives which each capture a different aspect of coherence, including whether salient spans of generated abstracts are hallucinated or appear in the input context, and the likelihood of sentence adjacency in generated abstracts. We measure the ability of Co-opNet to learn these objectives with arXiv scientific papers, using the abstracts as a proxy for gold long-form scientific article summaries. Empirical results from automatic and human evaluations demonstrate that Co-opNet learns to summarize with considerably improved global coherence compared to competitive baselines.
Biased associations have been a challenge in the development of classifiers for detecting toxic language, hindering both fairness and accuracy. As potential solutions, we investigate recently introduced debiasing methods for text classification datasets and models, as applied to toxic language detection. Our focus is on lexical (e.g., swear words, slurs, identity mentions) and dialectal markers (specifically African American English). Our comprehensive experiments establish that existing methods are limited in their ability to prevent biased behavior in current toxicity detectors. We then propose an automatic, dialect-aware data correction method, as a proof-of-concept. Despite the use of synthetic labels, this method reduces dialectal associations with toxicity. Overall, our findings show that debiasing a model trained on biased toxic language data is not as effective as simply relabeling the data to remove existing biases.
Scripts – prototypical event sequences describing everyday activities – have been shown to help understand narratives by providing expectations, resolving ambiguity, and filling in unstated information. However, to date they have proved hard to author or extract from text. In this work, we demonstrate for the first time that pre-trained neural language models can be finetuned to generate high-quality scripts, at varying levels of granularity, for a wide range of everyday scenarios (e.g., bake a cake). To do this, we collect a large (6.4k) crowdsourced partially ordered scripts (named proScript), that is substantially larger than prior datasets, and develop models that generate scripts by combining language generation and graph structure prediction. We define two complementary tasks: (i) edge prediction: given a scenario and unordered events, organize the events into a valid (possibly partial-order) script, and (ii) script generation: given only a scenario, generate events and organize them into a (possibly partial-order) script. Our experiments show that our models perform well (e.g., F1=75.7 on task (i)), illustrating a new approach to overcoming previous barriers to script collection. We also show that there is still significant room for improvement toward human level performance. Together, our tasks, dataset, and models offer a new research direction for learning script knowledge.
In social settings, much of human behavior is governed by unspoken rules of conduct rooted in societal norms. For artificial systems to be fully integrated into social environments, adherence to such norms is a central prerequisite. To investigate whether language generation models can serve as behavioral priors for systems deployed in social settings, we evaluate their ability to generate action descriptions that achieve predefined goals under normative constraints. Moreover, we examine if models can anticipate likely consequences of actions that either observe or violate known norms, or explain why certain actions are preferable by generating relevant norm hypotheses. For this purpose, we introduce Moral Stories, a crowd-sourced dataset of structured, branching narratives for the study of grounded, goal-oriented social reasoning. Finally, we propose decoding strategies that combine multiple expert models to significantly improve the quality of generated actions, consequences, and norms compared to strong baselines.
Contrastive explanations clarify why an event occurred in contrast to another. They are inherently intuitive to humans to both produce and comprehend. We propose a method to produce contrastive explanations in the latent space, via a projection of the input representation, such that only the features that differentiate two potential decisions are captured. Our modification allows model behavior to consider only contrastive reasoning, and uncover which aspects of the input are useful for and against particular decisions. Our contrastive explanations can additionally answer for which label, and against which alternative label, is a given input feature useful. We produce contrastive explanations via both high-level abstract concept attribution and low-level input token/span attribution for two NLP classification benchmarks. Our findings demonstrate the ability of label-contrastive explanations to provide fine-grained interpretability of model decisions.
Large language models have shown promising results in zero-shot settings. For example, they can perform multiple choice tasks simply by conditioning on a question and selecting the answer with the highest probability. However, ranking by string probability can be problematic due to surface form competition—wherein different surface forms compete for probability mass, even if they represent the same underlying concept in a given context, e.g. “computer” and “PC.” Since probability mass is finite, this lowers the probability of the correct answer, due to competition from other strings that are valid answers (but not one of the multiple choice options). We introduce Domain Conditional Pointwise Mutual Information, an alternative scoring function that directly compensates for surface form competition by simply reweighing each option according to its a priori likelihood within the context of a specific task. It achieves consistent gains in zero-shot performance over both calibrated and uncalibrated scoring functions on all GPT-2 and GPT-3 models on a variety of multiple choice datasets.
One of the challenges faced by conversational agents is their inability to identify unstated presumptions of their users’ commands, a task trivial for humans due to their common sense. In this paper, we propose a zero-shot commonsense reasoning system for conversational agents in an attempt to achieve this. Our reasoner uncovers unstated presumptions from user commands satisfying a general template of if-(state), then-(action), because-(goal). Our reasoner uses a state-of-the-art transformer-based generative commonsense knowledge base (KB) as its source of background knowledge for reasoning. We propose a novel and iterative knowledge query mechanism to extract multi-hop reasoning chains from the neural KB which uses symbolic logic rules to significantly reduce the search space. Similar to any KBs gathered to date, our commonsense KB is prone to missing knowledge. Therefore, we propose to conversationally elicit the missing knowledge from human users with our novel dynamic question generation strategy, which generates and presents contextualized queries to human users. We evaluate the model with a user study with human users that achieves a 35% higher success rate compared to SOTA.
Image captioning has conventionally relied on reference-based automatic evaluations, where machine captions are compared against captions written by humans. This is in contrast to the reference-free manner in which humans assess caption quality. In this paper, we report the surprising empirical finding that CLIP (Radford et al., 2021), a cross-modal model pretrained on 400M image+caption pairs from the web, can be used for robust automatic evaluation of image captioning without the need for references. Experiments spanning several corpora demonstrate that our new reference-free metric, CLIPScore, achieves the highest correlation with human judgements, outperforming existing reference-based metrics like CIDEr and SPICE. Information gain experiments demonstrate that CLIPScore, with its tight focus on image-text compatibility, is complementary to existing reference-based metrics that emphasize text-text similarities. Thus, we also present a reference-augmented version, RefCLIPScore, which achieves even higher correlation. Beyond literal description tasks, several case studies reveal domains where CLIPScore performs well (clip-art images, alt-text rating), but also where it is relatively weaker in comparison to reference-based metrics, e.g., news captions that require richer contextual knowledge.
Conditional text generation often requires lexical constraints, i.e., which words should or shouldn’t be included in the output text. While the dominant recipe for conditional text generation has been large-scale pretrained language models that are finetuned on the task-specific training data, such models do not learn to follow the underlying constraints reliably, even when supervised with large amounts of task-specific examples. We propose NeuroLogic Decoding, a simple yet effective algorithm that enables neural language models – supervised or not – to generate fluent text while satisfying complex lexical constraints. Our approach is powerful yet efficient. It handles any set of lexical constraints that is expressible under predicate logic, while its asymptotic runtime is equivalent to conventional beam search. Empirical results on four benchmarks show that NeuroLogic Decoding outperforms previous approaches, including algorithms that handle a subset of our constraints. Moreover, we find that unsupervised models with NeuroLogic Decoding often outperform supervised models with conventional decoding, even when the latter is based on considerably larger networks. Our results suggest the limit of large-scale neural networks for fine-grained controllable generation and the promise of inference-time algorithms.
Natural language inference requires reasoning about contradictions, negations, and their commonsense implications. Given a simple premise (e.g., “I’m mad at you”), humans can reason about the varying shades of contradictory statements ranging from straightforward negations (“I’m not mad at you”) to commonsense contradictions (“I’m happy”). Moreover, these negated or contradictory statements shift the commonsense implications of the original premise in interesting and nontrivial ways. For example, while “I’m mad” implies “I’m unhappy about something,” negating the premise does not necessarily negate the corresponding commonsense implications. In this paper, we present the first comprehensive study focusing on commonsense implications of negated statements and contradictions. We introduce ANION, a new commonsense knowledge graph with 624K if-then rules focusing on negated and contradictory events. We then present joint generative and discriminative inference models for this new resource, providing novel empirical insights on how logical negations and commonsense contradictions reshape the commonsense implications of their original premises.
We propose TuringAdvice, a new challenge task and dataset for language understanding models. Given a written situation that a real person is currently facing, a model must generate helpful advice in natural language. Our evaluation framework tests a fundamental aspect of human language understanding: our ability to use language to resolve open-ended situations by communicating with each other. Empirical results show that today’s models struggle at TuringAdvice, even multibillion parameter models finetuned on 600k in-domain training examples. The best model, T5, writes advice that is at least as helpful as human-written advice in only 14% of cases; a much larger non-finetunable GPT3 model does even worse at 4%. This low performance reveals language understanding errors that are hard to spot outside of a generative setting, showing much room for progress.
Publicly available, large pretrained Language Models (LMs) generate text with remarkable quality, but only sequentially from left to right. As a result, they are not immediately applicable to generation tasks that break the unidirectional assumption, such as paraphrasing or text-infilling, necessitating task-specific supervision. In this paper, we present Reflective Decoding, a novel unsupervised algorithm that allows for direct application of unidirectional LMs to non-sequential tasks. Our 2-step approach requires no supervision or even parallel corpora, only two off-the-shelf pretrained LMs in opposite directions: forward and backward. First, in the contextualization step, we use LMs to generate ensembles of past and future contexts which collectively capture the input (e.g. the source sentence for paraphrasing). Second, in the reflection step, we condition on these “context ensembles”, generating outputs that are compatible with them. Comprehensive empirical results demonstrate that Reflective Decoding outperforms strong unsupervised baselines on both paraphrasing and abductive text infilling, significantly narrowing the gap between unsupervised and supervised methods. Reflective Decoding surpasses multiple supervised baselines on various metrics including human evaluation.
Understanding manipulated media, from automatically generated ‘deepfakes’ to manually edited ones, raises novel research challenges. Because the vast majority of edited or manipulated images are benign, such as photoshopped images for visual enhancements, the key challenge is to understand the complex layers of underlying intents of media edits and their implications with respect to disinformation. In this paper, we study Edited Media Frames, a new formalism to understand visual media manipulation as structured annotations with respect to the intents, emotional reactions, attacks on individuals, and the overall implications of disinformation. We introduce a dataset for our task, EMU, with 56k question-answer pairs written in rich natural language. We evaluate a wide variety of vision-and-language models for our task, and introduce a new model PELICAN, which builds upon recent progress in pretrained multimodal representations. Our model obtains promising results on our dataset, with humans rating its answers as accurate 48.2% of the time. At the same time, there is still much work to be done – and we provide analysis that highlights areas for further progress.
We propose PIGLeT: a model that learns physical commonsense knowledge through interaction, and then uses this knowledge to ground language. We factorize PIGLeT into a physical dynamics model, and a separate language model. Our dynamics model learns not just what objects are but also what they do: glass cups break when thrown, plastic ones don’t. We then use it as the interface to our language model, giving us a unified model of linguistic form and grounded meaning. PIGLeT can read a sentence, simulate neurally what might happen next, and then communicate that result through a literal symbolic representation, or natural language. Experimental results show that our model effectively learns world dynamics, along with how to communicate them. It is able to correctly forecast what happens next given an English sentence over 80% of the time, outperforming a 100x larger, text-to-text approach by over 10%. Likewise, its natural language summaries of physical interactions are also judged by humans as more accurate than LM alternatives. We present comprehensive analysis showing room for future work.
Despite recent advances in natural language generation, it remains challenging to control attributes of generated text. We propose DExperts: Decoding-time Experts, a decoding-time method for controlled text generation that combines a pretrained language model with “expert” LMs and/or “anti-expert” LMs in a product of experts. Intuitively, under the ensemble, tokens only get high probability if they are considered likely by the experts, and unlikely by the anti-experts. We apply DExperts to language detoxification and sentiment-controlled generation, where we outperform existing controllable generation methods on both automatic and human evaluations. Moreover, because DExperts operates only on the output of the pretrained LM, it is effective with (anti-)experts of smaller size, including when operating on GPT-3. Our work highlights the promise of tuning small LMs on text with (un)desirable attributes for efficient decoding-time steering.
Everyday conversations require understanding everyday events, which in turn, requires understanding temporal commonsense concepts interwoven with those events. Despite recent progress with massive pre-trained language models (LMs) such as T5 and GPT-3, their capability of temporal reasoning in dialogs remains largely under-explored. In this paper, we present the first study to investigate pre-trained LMs for their temporal reasoning capabilities in dialogs by introducing a new task and a crowd-sourced English challenge set, TimeDial. We formulate TimeDial as a multiple choice cloze task with over 1.1K carefully curated dialogs. Empirical results demonstrate that even the best performing models struggle on this task compared to humans, with 23 absolute points of gap in accuracy. Furthermore, our analysis reveals that the models fail to reason about dialog context correctly; instead, they rely on shallow cues based on existing temporal patterns in context, motivating future research for modeling temporal concepts in text and robust contextual reasoning about them. The dataset is publicly available at https://github.com/google-research-datasets/timedial.
Social norms—the unspoken commonsense rules about acceptable social behavior—are crucial in understanding the underlying causes and intents of people’s actions in narratives. For example, underlying an action such as “wanting to call cops on my neighbor” are social norms that inform our conduct, such as “It is expected that you report crimes.” We present SOCIAL CHEMISTRY, a new conceptual formalism to study people’s everyday social norms and moral judgments over a rich spectrum of real life situations described in natural language. We introduce SOCIAL-CHEM-101, a large-scale corpus that catalogs 292k rules-of-thumb such as “It is rude to run a blender at 5am” as the basic conceptual units. Each rule-of-thumb is further broken down with 12 different dimensions of people’s judgments, including social judgments of good and bad, moral foundations, expected cultural pressure, and assumed legality, which together amount to over 4.5 million annotations of categorical labels and free-text descriptions. Comprehensive empirical results based on state-of-the-art neural models demonstrate that computational modeling of social norms is a promising research direction. Our model framework, Neural Norm Transformer, learns and generalizes SOCIAL-CHEM-101 to successfully reason about previously unseen situations, generating relevant (and potentially novel) attribute-aware social rules-of-thumb.
Abductive and counterfactual reasoning, core abilities of everyday human cognition, require reasoning about what might have happened at time t, while conditioning on multiple contexts from the relative past and future. However, simultaneous incorporation of past and future contexts using generative language models (LMs) can be challenging, as they are trained either to condition only on the past context or to perform narrowly scoped text-infilling. In this paper, we propose DeLorean, a new unsupervised decoding algorithm that can flexibly incorporate both the past and future contexts using only off-the-shelf, left-to-right language models and no supervision. The key intuition of our algorithm is incorporating the future through back-propagation, during which, we only update the internal representation of the output while fixing the model parameters. By alternating between forward and backward propagation, DeLorean can decode the output representation that reflects both the left and right contexts. We demonstrate that our approach is general and applicable to two nonmonotonic reasoning tasks: abductive text generation and counterfactual story revision, where DeLorean outperforms a range of unsupervised and some supervised methods, based on automatic and human evaluation.
We propose the task of outline-conditioned story generation: given an outline as a set of phrases that describe key characters and events to appear in a story, the task is to generate a coherent narrative that is consistent with the provided outline. This task is challenging as the input only provides a rough sketch of the plot, and thus, models need to generate a story by interweaving the key points provided in the outline. This requires the model to keep track of the dynamic states of the latent plot, conditioning on the input outline while generating the full story. We present PlotMachines, a neural narrative model that learns to transform an outline into a coherent story by tracking the dynamic plot states. In addition, we enrich PlotMachines with high-level discourse structure so that the model can learn different writing styles corresponding to different parts of the narrative. Comprehensive experiments over three fiction and non-fiction datasets demonstrate that large-scale language models, such as GPT-2 and Grover, despite their impressive generation performance, are not sufficient in generating coherent narratives for the given outline, and dynamic plot state tracking is important for composing narratives with tighter, more consistent plots.
Natural language understanding involves reading between the lines with implicit background knowledge. Current systems either rely on pre-trained language models as the sole implicit source of world knowledge, or resort to external knowledge bases (KBs) to incorporate additional relevant knowledge. We propose an unsupervised framework based on self-talk as a novel alternative to multiple-choice commonsense tasks. Inspired by inquiry-based discovery learning (Bruner, 1961), our approach inquires language models with a number of information seeking questions such as “what is the definition of...” to discover additional background knowledge. Empirical results demonstrate that the self-talk procedure substantially improves the performance of zero-shot language model baselines on four out of six commonsense benchmarks, and competes with models that obtain knowledge from external KBs. While our approach improves performance on several benchmarks, the self-talk induced knowledge even when leading to correct answers is not always seen as helpful by human judges, raising interesting questions about the inner-workings of pre-trained language models for commonsense reasoning.
Unconscious biases continue to be prevalent in modern text and media, calling for algorithms that can assist writers with bias correction. For example, a female character in a story is often portrayed as passive and powerless (“_She daydreams about being a doctor_”) while a man is portrayed as more proactive and powerful (“_He pursues his dream of being a doctor_”). We formulate **Controllable Debiasing**, a new revision task that aims to rewrite a given text to correct the implicit and potentially undesirable bias in character portrayals. We then introduce PowerTransformer as an approach that debiases text through the lens of connotation frames (Sap et al., 2017), which encode pragmatic knowledge of implied power dynamics with respect to verb predicates. One key challenge of our task is the lack of parallel corpora. To address this challenge, we adopt an unsupervised approach using auxiliary supervision with related tasks such as paraphrasing and self-supervision based on a reconstruction loss, building on pretrained language models. Through comprehensive experiments based on automatic and human evaluations, we demonstrate that our approach outperforms ablations and existing methods from related tasks. Furthermore, we demonstrate the use of PowerTransformer as a step toward mitigating the well-documented gender bias in character portrayal in movie scripts.
Large datasets have become commonplace in NLP research. However, the increased emphasis on data quantity has made it challenging to assess the quality of data. We introduce Data Maps—a model-based tool to characterize and diagnose datasets. We leverage a largely ignored source of information: the behavior of the model on individual instances during training (training dynamics) for building data maps. This yields two intuitive measures for each example—the model’s confidence in the true class, and the variability of this confidence across epochs—obtained in a single run of training. Experiments on four datasets show that these model-dependent measures reveal three distinct regions in the data map, each with pronounced characteristics. First, our data maps show the presence of “ambiguous” regions with respect to the model, which contribute the most towards out-of-distribution generalization. Second, the most populous regions in the data are “easy to learn” for the model, and play an important role in model optimization. Finally, data maps uncover a region with instances that the model finds “hard to learn”; these often correspond to labeling errors. Our results indicate that a shift in focus from quantity to quality of data could lead to robust models and improved out-of-distribution generalization.
Recent advances in commonsense reasoning depend on large-scale human-annotated training sets to achieve peak performance. However, manual curation of training sets is expensive and has been shown to introduce annotation artifacts that neural models can readily exploit and overfit to. We propose a novel generative data augmentation technique, G-DAUGˆC, that aims to achieve more accurate and robust learning in a low-resource setting. Our approach generates synthetic examples using pretrained language models and selects the most informative and diverse set of examples for data augmentation. On experiments with multiple commonsense reasoning benchmarks, G-DAUGˆC consistently outperforms existing data augmentation methods based on back-translation, establishing a new state-of-the-art on WinoGrande, CODAH, and CommonsenseQA, as well as enhances out-of-distribution generalization, proving to be robust against adversaries or perturbations. Our analysis demonstrates that G-DAUGˆC produces a diverse set of fluent training examples, and that its selection and training approaches are important for performance.
Recently, large-scale pre-trained language models have demonstrated impressive performance on several commonsense-reasoning benchmark datasets. However, building machines with commonsense to compose realistically plausible sentences remains challenging. In this paper, we present a constrained text generation task, CommonGen associated with a benchmark dataset, to explicitly test machines for the ability of generative commonsense reasoning. Given a set of common concepts (e.g., dog, frisbee, catch, throw); the task is to generate a coherent sentence describing an everyday scenario using these concepts (e.g., “a man throws a frisbee and his dog catches it”). The CommonGen task is challenging because it inherently requires 1) relational reasoning with background commonsense knowledge and 2) compositional generalization ability to work on unseen concept combinations. Our dataset, constructed through a combination of crowdsourced and existing caption corpora, consists of 77k commonsense descriptions over 35k unique concept-sets. Experiments show that there is a large gap between state-of-the-art text generation models (e.g., T5) and human performance (31.6% v.s. 63.5% in SPICE metric). Furthermore, we demonstrate that the learned generative commonsense reasoning capability can be transferred to improve downstream tasks such as CommonsenseQA (76.9% to 78.4 in dev accuracy) by generating additional context.
Natural language rationales could provide intuitive, higher-level explanations that are easily understandable by humans, complementing the more broadly studied lower-level explanations based on gradients or attention weights. We present the first study focused on generating natural language rationales across several complex visual reasoning tasks: visual commonsense reasoning, visual-textual entailment, and visual question answering. The key challenge of accurate rationalization is comprehensive image understanding at all levels: not just their explicit content at the pixel level, but their contextual contents at the semantic and pragmatic levels. We present RationaleˆVT Transformer, an integrated model that learns to generate free-text rationales by combining pretrained language models with object recognition, grounded visual semantic frames, and visual commonsense graphs. Our experiments show that free-text rationalization is a promising research direction to complement model interpretability for complex visual-textual reasoning tasks. In addition, we find that integration of richer semantic and pragmatic visual features improves visual fidelity of rationales.
Pretrained neural language models (LMs) are prone to generating racist, sexist, or otherwise toxic language which hinders their safe deployment. We investigate the extent to which pretrained LMs can be prompted to generate toxic language, and the effectiveness of controllable text generation algorithms at preventing such toxic degeneration. We create and release RealToxicityPrompts, a dataset of 100K naturally occurring, sentence-level prompts derived from a large corpus of English web text, paired with toxicity scores from a widely-used toxicity classifier. Using RealToxicityPrompts, we find that pretrained LMs can degenerate into toxic text even from seemingly innocuous prompts. We empirically assess several controllable generation methods, and find that while data- or compute-intensive methods (e.g., adaptive pretraining on non-toxic data) are more effective at steering away from toxicity than simpler solutions (e.g., banning “bad” words), no current method is failsafe against neural toxic degeneration. To pinpoint the potential cause of such persistent toxic degeneration, we analyze two web text corpora used to pretrain several LMs (including GPT-2; Radford et. al, 2019), and find a significant amount of offensive, factually unreliable, and otherwise toxic content. Our work provides a test bed for evaluating toxic generations by LMs and stresses the need for better data selection processes for pretraining.
Defeasible inference is a mode of reasoning in which an inference (X is a bird, therefore X flies) may be weakened or overturned in light of new evidence (X is a penguin). Though long recognized in classical AI and philosophy, defeasible inference has not been extensively studied in the context of contemporary data-driven research on natural language inference and commonsense reasoning. We introduce Defeasible NLI (abbreviated 𝛿-NLI), a dataset for defeasible inference in natural language. Defeasible NLI contains extensions to three existing inference datasets covering diverse modes of reasoning: common sense, natural language inference, and social norms. From Defeasible NLI, we develop both a classification and generation task for defeasible inference, and demonstrate that the generation task is much more challenging. Despite lagging human performance, however, generative models trained on this data are capable of writing sentences that weaken or strengthen a specified inference up to 68% of the time.
We investigate the use of NLP as a measure of the cognitive processes involved in storytelling, contrasting imagination and recollection of events. To facilitate this, we collect and release Hippocorpus, a dataset of 7,000 stories about imagined and recalled events. We introduce a measure of narrative flow and use this to examine the narratives for imagined and recalled events. Additionally, we measure the differential recruitment of knowledge attributed to semantic memory versus episodic memory (Tulving, 1972) for imagined and recalled storytelling by comparing the frequency of descriptions of general commonsense events with more specific realis events. Our analyses show that imagined stories have a substantially more linear narrative flow, compared to recalled stories in which adjacent sentences are more disconnected. In addition, while recalled stories rely more on autobiographical events based on episodic memory, imagined stories express more commonsense knowledge based on semantic memory. Finally, our measures reveal the effect of narrativization of memories in stories (e.g., stories about frequently recalled memories flow more linearly; Bartlett, 1932). Our findings highlight the potential of using NLP tools to study the traces of human cognition in language.
Warning: this paper contains content that may be offensive or upsetting. Language has the power to reinforce stereotypes and project social biases onto others. At the core of the challenge is that it is rarely what is stated explicitly, but rather the implied meanings, that frame people’s judgments about others. For example, given a statement that “we shouldn’t lower our standards to hire more women,” most listeners will infer the implicature intended by the speaker - that “women (candidates) are less qualified.” Most semantic formalisms, to date, do not capture such pragmatic implications in which people express social biases and power differentials in language. We introduce Social Bias Frames, a new conceptual formalism that aims to model the pragmatic frames in which people project social biases and stereotypes onto others. In addition, we introduce the Social Bias Inference Corpus to support large-scale modelling and evaluation with 150k structured annotations of social media posts, covering over 34k implications about a thousand demographic groups. We then establish baseline approaches that learn to recover Social Bias Frames from unstructured text. We find that while state-of-the-art neural models are effective at high-level categorization of whether a given statement projects unwanted social bias (80% F1), they are not effective at spelling out more detailed explanations in terms of Social Bias Frames. Our study motivates future work that combines structured pragmatic inference with commonsense reasoning on social implications.
Commonsense knowledge, such as knowing that “bumping into people annoys them” or “rain makes the road slippery”, helps humans navigate everyday situations seamlessly. Yet, endowing machines with such human-like commonsense reasoning capabilities has remained an elusive goal of artificial intelligence research for decades. In recent years, commonsense knowledge and reasoning have received renewed attention from the natural language processing (NLP) community, yielding exploratory studies in automated commonsense understanding. We organize this tutorial to provide researchers with the critical foundations and recent advances in commonsense representation and reasoning, in the hopes of casting a brighter light on this promising area of future research. In our tutorial, we will (1) outline the various types of commonsense (e.g., physical, social), and (2) discuss techniques to gather and represent commonsense knowledge, while highlighting the challenges specific to this type of knowledge (e.g., reporting bias). We will then (3) discuss the types of commonsense knowledge captured by modern NLP systems (e.g., large pretrained language models), and (4) present ways to measure systems’ commonsense reasoning abilities. We will finish with (5) a discussion of various ways in which commonsense reasoning can be used to improve performance on NLP tasks, exemplified by an (6) interactive session on integrating commonsense into a downstream task.
Mining commonsense knowledge from corpora suffers from reporting bias, over-representing the rare at the expense of the trivial (Gordon and Van Durme, 2013). We study to what extent pre-trained language models overcome this issue. We find that while their generalization capacity allows them to better estimate the plausibility of frequent but unspoken of actions, outcomes, and properties, they also tend to overestimate that of the very rare, amplifying the bias that already exists in their training corpus.
We investigate how annotators’ insensitivity to differences in dialect can lead to racial bias in automatic hate speech detection models, potentially amplifying harm against minority populations. We first uncover unexpected correlations between surface markers of African American English (AAE) and ratings of toxicity in several widely-used hate speech datasets. Then, we show that models trained on these corpora acquire and propagate these biases, such that AAE tweets and tweets by self-identified African Americans are up to two times more likely to be labelled as offensive compared to others. Finally, we propose *dialect* and *race priming* as ways to reduce the racial bias in annotation, showing that when annotators are made explicitly aware of an AAE tweet’s dialect they are significantly less likely to label the tweet as offensive.
We present the first comprehensive study on automatic knowledge base construction for two prevalent commonsense knowledge graphs: ATOMIC (Sap et al., 2019) and ConceptNet (Speer et al., 2017). Contrary to many conventional KBs that store knowledge with canonical templates, commonsense KBs only store loosely structured open-text descriptions of knowledge. We posit that an important step toward automatic commonsense completion is the development of generative models of commonsense knowledge, and propose COMmonsEnse Transformers (COMET) that learn to generate rich and diverse commonsense descriptions in natural language. Despite the challenges of commonsense modeling, our investigation reveals promising results when implicit knowledge from deep pre-trained language models is transferred to generate explicit knowledge in commonsense knowledge graphs. Empirical results demonstrate that COMET is able to generate novel knowledge that humans rate as high quality, with up to 77.5% (ATOMIC) and 91.7% (ConceptNet) precision at top 1, which approaches human performance for these resources. Our findings suggest that using generative commonsense models for automatic commonsense KB completion could soon be a plausible alternative to extractive methods.
Recent work by Zellers et al. (2018) introduced a new task of commonsense natural language inference: given an event description such as “A woman sits at a piano,” a machine must select the most likely followup: “She sets her fingers on the keys.” With the introduction of BERT, near human-level performance was reached. Does this mean that machines can perform human level commonsense inference? In this paper, we show that commonsense inference still proves difficult for even state-of-the-art models, by presenting HellaSwag, a new challenge dataset. Though its questions are trivial for humans (>95% accuracy), state-of-the-art models struggle (<48%). We achieve this via Adversarial Filtering (AF), a data collection paradigm wherein a series of discriminators iteratively select an adversarial set of machine-generated wrong answers. AF proves to be surprisingly robust. The key insight is to scale up the length and complexity of the dataset examples towards a critical ‘Goldilocks’ zone wherein generated text is ridiculous to humans, yet often misclassified by state-of-the-art models. Our construction of HellaSwag, and its resulting difficulty, sheds light on the inner workings of deep pretrained models. More broadly, it suggests a new path forward for NLP research, in which benchmarks co-evolve with the evolving state-of-the-art in an adversarial way, so as to present ever-harder challenges.
Although neural conversational models are effective in learning how to produce fluent responses, their primary challenge lies in knowing what to say to make the conversation contentful and non-vacuous. We present a new end-to-end approach to contentful neural conversation that jointly models response generation and on-demand machine reading. The key idea is to provide the conversation model with relevant long-form text on the fly as a source of external knowledge. The model performs QA-style reading comprehension on this text in response to each conversational turn, thereby allowing for more focused integration of external knowledge than has been possible in prior approaches. To support further research on knowledge-grounded conversation, we introduce a new large-scale conversation dataset grounded in external web pages (2.8M turns, 7.4M sentences of grounding). Both human evaluation and automated metrics show that our approach results in more contentful responses compared to a variety of previous methods, improving both the informativeness and diversity of generated output.
We present DREAM, the first dialogue-based multiple-choice reading comprehension data set. Collected from English as a Foreign Language examinations designed by human experts to evaluate the comprehension level of Chinese learners of English, our data set contains 10,197 multiple-choice questions for 6,444 dialogues. In contrast to existing reading comprehension data sets, DREAM is the first to focus on in-depth multi-turn multi-party dialogue understanding. DREAM is likely to present significant challenges for existing reading comprehension systems: 84% of answers are non-extractive, 85% of questions require reasoning beyond a single sentence, and 34% of questions also involve commonsense knowledge. We apply several popular neural reading comprehension models that primarily exploit surface information within the text and find them to, at best, just barely outperform a rule-based approach. We next investigate the effects of incorporating dialogue structure and different kinds of general world knowledge into both rule-based and (neural and non-neural) machine learning-based reading comprehension models. Experimental results on the DREAM data set show the effectiveness of dialogue structure and general world knowledge. DREAM is available at https://dataset.org/dream/.
We introduce a large-scale dataset of math word problems and an interpretable neural math problem solver by learning to map problems to their operation programs. Due to annotation challenges, current datasets in this domain have been either relatively small in scale or did not offer precise operational annotations over diverse problem types. We introduce a new representation language to model operation programs corresponding to each math problem that aim to improve both the performance and the interpretability of the learned models. Using this representation language, we significantly enhance the AQUA-RAT dataset with fully-specified operational programs. We additionally introduce a neural sequence-to-program model with automatic problem categorization. Our experiments show improvements over competitive baselines in our dataset as well as the AQUA-RAT dataset. The results are still lower than human performance indicating that the dataset poses new challenges for future research. Our dataset is available at https://math-qa.github.io/math-QA/
Understanding procedural language requires reasoning about both hierarchical and temporal relations between events. For example, “boiling pasta” is a sub-event of “making a pasta dish”, typically happens before “draining pasta,” and requires the use of omitted tools (e.g. a strainer, sink...). While people are able to choose when and how to use abstract versus concrete instructions, the NLP community lacks corpora and tasks for evaluating if our models can do the same. In this paper, we introduce KidsCook, a parallel script corpus, as well as a cloze task which matches video captions with missing procedural details. Experimental results show that state-of-the-art models struggle at this task, which requires inducing functional commonsense knowledge not explicitly stated in text.
Core to the vision-and-language navigation (VLN) challenge is building robust instruction representations and action decoding schemes, which can generalize well to previously unseen instructions and environments. In this paper, we report two simple but highly effective methods to address these challenges and lead to a new state-of-the-art performance. First, we adapt large-scale pretrained language models to learn text representations that generalize better to previously unseen instructions. Second, we propose a stochastic sampling scheme to reduce the considerable gap between the expert actions in training and sampled actions in test, so that the agent can learn to correct its own mistakes during long sequential action decoding. Combining the two techniques, we achieve a new state of the art on the Room-to-Room benchmark with 6% absolute gain over the previous best result (47% -> 53%) on the Success Rate weighted by Path Length metric.
Understanding narratives requires reading between the lines, which in turn, requires interpreting the likely causes and effects of events, even when they are not mentioned explicitly. In this paper, we introduce Cosmos QA, a large-scale dataset of 35,600 problems that require commonsense-based reading comprehension, formulated as multiple-choice questions. In stark contrast to most existing reading comprehension datasets where the questions focus on factual and literal understanding of the context paragraph, our dataset focuses on reading between the lines over a diverse collection of people’s everyday narratives, asking such questions as “what might be the possible reason of ...?", or “what would have happened if ..." that require reasoning beyond the exact text spans in the context. To establish baseline performances on Cosmos QA, we experiment with several state-of-the-art neural architectures for reading comprehension, and also propose a new architecture that improves over the competitive baselines. Experimental results demonstrate a significant gap between machine (68.4%) and human performance (94%), pointing to avenues for future research on commonsense machine comprehension. Dataset, code and leaderboard is publicly available at https://wilburone.github.io/cosmos.
The principle of the Information Bottleneck (Tishby et al., 1999) produces a summary of information X optimized to predict some other relevant information Y. In this paper, we propose a novel approach to unsupervised sentence summarization by mapping the Information Bottleneck principle to a conditional language modelling objective: given a sentence, our approach seeks a compressed sentence that can best predict the next sentence. Our iterative algorithm under the Information Bottleneck objective searches gradually shorter subsequences of the given sentence while maximizing the probability of the next sentence conditioned on the summary. Using only pretrained language models with no direct supervision, our approach can efficiently perform extractive sentence summarization over a large corpus. Building on our unsupervised extractive summarization, we also present a new approach to self-supervised abstractive summarization, where a transformer-based language model is trained on the output summaries of our unsupervised method. Empirical results demonstrate that our extractive method outperforms other unsupervised models on multiple automatic metrics. In addition, we find that our self-supervised abstractive model outperforms unsupervised baselines (including our own) by human evaluation along multiple attributes.
We introduce Social IQa, the first large-scale benchmark for commonsense reasoning about social situations. Social IQa contains 38,000 multiple choice questions for probing emotional and social intelligence in a variety of everyday situations (e.g., Q: “Jordan wanted to tell Tracy a secret, so Jordan leaned towards Tracy. Why did Jordan do this?” A: “Make sure no one else could hear”). Through crowdsourcing, we collect commonsense questions along with correct and incorrect answers about social interactions, using a new framework that mitigates stylistic artifacts in incorrect answers by asking workers to provide the right answer to a different but related question. Empirical results show that our benchmark is challenging for existing question-answering models based on pretrained language models, compared to human performance (>20% gap). Notably, we further establish Social IQa as a resource for transfer learning of commonsense knowledge, achieving state-of-the-art performance on multiple commonsense reasoning tasks (Winograd Schemas, COPA).
Counterfactual reasoning requires predicting how alternative events, contrary to what actually happened, might have resulted in different outcomes. Despite being considered a necessary component of AI-complete systems, few resources have been developed for evaluating counterfactual reasoning in narratives. In this paper, we propose Counterfactual Story Rewriting: given an original story and an intervening counterfactual event, the task is to minimally revise the story to make it compatible with the given counterfactual event. Solving this task will require deep understanding of causal narrative chains and counterfactual invariance, and integration of such story reasoning capabilities into conditional language generation models. We present TIMETRAVEL, a new dataset of 29,849 counterfactual rewritings, each with the original story, a counterfactual event, and human-generated revision of the original story compatible with the counterfactual event. Additionally, we include 81,407 counterfactual “branches” without a rewritten storyline to support future work on semi- or un-supervised approaches to counterfactual story rewriting. Finally, we evaluate the counterfactual rewriting capacities of several competitive baselines based on pretrained language models, and assess whether common overlap and model-based automatic metrics for text generation correlate well with human scores for counterfactual rewriting.
Given a partial description like “she opened the hood of the car,” humans can reason about the situation and anticipate what might come next (”then, she examined the engine”). In this paper, we introduce the task of grounded commonsense inference, unifying natural language inference and commonsense reasoning. We present SWAG, a new dataset with 113k multiple choice questions about a rich spectrum of grounded situations. To address the recurring challenges of the annotation artifacts and human biases found in many existing datasets, we propose Adversarial Filtering (AF), a novel procedure that constructs a de-biased dataset by iteratively training an ensemble of stylistic classifiers, and using them to filter the data. To account for the aggressive adversarial filtering, we use state-of-the-art language models to massively oversample a diverse set of potential counterfactuals. Empirical results demonstrate that while humans can solve the resulting inference problems with high accuracy (88%), various competitive models struggle on our task. We provide comprehensive analysis that indicates significant opportunities for future research.
We present end-to-end neural models for detecting metaphorical word use in context. We show that relatively standard BiLSTM models which operate on complete sentences work well in this setting, in comparison to previous work that used more restricted forms of linguistic context. These models establish a new state-of-the-art on existing verb metaphor detection benchmarks, and show strong performance on jointly predicting the metaphoricity of all words in a running text.
We present QuAC, a dataset for Question Answering in Context that contains 14K information-seeking QA dialogs (100K questions in total). The dialogs involve two crowd workers: (1) a student who poses a sequence of freeform questions to learn as much as possible about a hidden Wikipedia text, and (2) a teacher who answers the questions by providing short excerpts from the text. QuAC introduces challenges not found in existing machine comprehension datasets: its questions are often more open-ended, unanswerable, or only meaningful within the dialog context, as we show in a detailed qualitative evaluation. We also report results for a number of reference models, including a recently state-of-the-art reading comprehension architecture extended to model dialog context. Our best model underperforms humans by 20 F1, suggesting that there is significant room for future work on this data. Dataset, baseline, and leaderboard available at http://quac.ai.
In this paper, we investigate the use of discourse-aware rewards with reinforcement learning to guide a model to generate long, coherent text. In particular, we propose to learn neural rewards to model cross-sentence ordering as a means to approximate desired discourse structure. Empirical results demonstrate that a generator trained with the learned reward produces more coherent and less repetitive text than models trained with cross-entropy or with reinforcement learning with commonly used scores as rewards.
We present deep communicating agents in an encoder-decoder architecture to address the challenges of representing a long document for abstractive summarization. With deep communicating agents, the task of encoding a long text is divided across multiple collaborating agents, each in charge of a subsection of the input text. These encoders are connected to a single decoder, trained end-to-end using reinforcement learning to generate a focused and coherent summary. Empirical results demonstrate that multiple communicating encoders lead to a higher quality summary compared to several strong baselines, including those based on a single encoder or multiple non-communicating encoders.
We present the first neural poetry translation system. Unlike previous works that often fail to produce any translation for fixed rhyme and rhythm patterns, our system always translates a source text to an English poem. Human evaluation of the translations ranks the quality as acceptable 78.2% of the time.
We present Sounding Board, a social chatbot that won the 2017 Amazon Alexa Prize. The system architecture consists of several components including spoken language processing, dialogue management, language generation, and content management, with emphasis on user-centric and content-driven design. We also share insights gained from large-scale online logs based on 160,000 conversations with real-world users.
We introduce a new entity typing task: given a sentence with an entity mention, the goal is to predict a set of free-form phrases (e.g. skyscraper, songwriter, or criminal) that describe appropriate types for the target entity. This formulation allows us to use a new type of distant supervision at large scale: head words, which indicate the type of the noun phrases they appear in. We show that these ultra-fine types can be crowd-sourced, and introduce new evaluation sets that are much more diverse and fine-grained than existing benchmarks. We present a model that can predict ultra-fine types, and is trained using a multitask objective that pools our new head-word supervision with prior supervision from entity linking. Experimental results demonstrate that our model is effective in predicting entity types at varying granularity; it achieves state of the art performance on an existing fine-grained entity typing benchmark, and sets baselines for our newly-introduced datasets.
We investigate a new commonsense inference task: given an event described in a short free-form text (“X drinks coffee in the morning”), a system reasons about the likely intents (“X wants to stay awake”) and reactions (“X feels alert”) of the event’s participants. To support this study, we construct a new crowdsourced corpus of 25,000 event phrases covering a diverse range of everyday events and situations. We report baseline performance on this task, demonstrating that neural encoder-decoder models can successfully compose embedding representations of previously unseen events and reason about the likely intents and reactions of the event participants. In addition, we demonstrate how commonsense inference on people’s intents and reactions can help unveil the implicit gender inequality prevalent in modern movie scripts.
Despite their local fluency, long-form text generated from RNNs is often generic, repetitive, and even self-contradictory. We propose a unified learning framework that collectively addresses all the above issues by composing a committee of discriminators that can guide a base RNN generator towards more globally coherent generations. More concretely, discriminators each specialize in a different principle of communication, such as Grice’s maxims, and are collectively combined with the base RNN generator through a composite decoding objective. Human evaluation demonstrates that text generated by our model is preferred over that of baselines by a large margin, significantly enhancing the overall coherence, style, and information of the generations.
Understanding a narrative requires reading between the lines and reasoning about the unspoken but obvious implications about events and people’s mental states — a capability that is trivial for humans but remarkably hard for machines. To facilitate research addressing this challenge, we introduce a new annotation framework to explain naive psychology of story characters as fully-specified chains of mental states with respect to motivations and emotional reactions. Our work presents a new large-scale dataset with rich low-level annotations and establishes baseline performance on several new tasks, suggesting avenues for future research.
This paper describes University of Washington NLP’s submission for the Linking Models of Lexical, Sentential and Discourse-level Semantics (LSDSem 2017) shared task—the Story Cloze Task. Our system is a linear classifier with a variety of features, including both the scores of a neural language model and style features. We report 75.2% accuracy on the task. A further discussion of our results can be found in Schwartz et al. (2017).
Sequence-to-sequence models have shown strong performance across a broad range of applications. However, their application to parsing and generating text using Abstract Meaning Representation (AMR) has been limited, due to the relatively limited amount of labeled data and the non-sequential nature of the AMR graphs. We present a novel training procedure that can lift this limitation using millions of unlabeled sentences and careful preprocessing of the AMR graphs. For AMR parsing, our model achieves competitive results of 62.1 SMATCH, the current best score reported without significant use of external semantic resources. For AMR generation, our model establishes a new state-of-the-art performance of BLEU 33.8. We present extensive ablative and qualitative analysis including strong evidence that sequence-based AMR models are robust against ordering variations of graph-to-sequence conversions.
Learning commonsense knowledge from natural language text is nontrivial due to reporting bias: people rarely state the obvious, e.g., “My house is bigger than me.” However, while rarely stated explicitly, this trivial everyday knowledge does influence the way people talk about the world, which provides indirect clues to reason about the world. For example, a statement like, “Tyler entered his house” implies that his house is bigger than Tyler. In this paper, we present an approach to infer relative physical knowledge of actions and objects along five dimensions (e.g., size, weight, and strength) from unstructured natural language text. We frame knowledge acquisition as joint inference over two closely related problems: learning (1) relative physical knowledge of object pairs and (2) physical implications of actions when applied to those object pairs. Empirical results demonstrate that it is possible to extract knowledge of actions and objects from language and that joint inference over different types of knowledge improves performance.
People around the globe respond to major real world events through social media. To study targeted public sentiments across many languages and geographic locations, we introduce multilingual connotation frames: an extension from English connotation frames of Rashkin et al. (2016) with 10 additional European languages, focusing on the implied sentiments among event participants engaged in a frame. As a case study, we present large scale analysis on targeted public sentiments toward salient events and entities using 1.2 million multilingual connotation frames extracted from Twitter.
A writer’s style depends not just on personal traits but also on her intent and mental state. In this paper, we show how variants of the same writing task can lead to measurable differences in writing style. We present a case study based on the story cloze task (Mostafazadeh et al., 2016a), where annotators were assigned similar writing tasks with different constraints: (1) writing an entire story, (2) adding a story ending for a given story context, and (3) adding an incoherent ending to a story. We show that a simple linear classifier informed by stylistic features is able to successfully distinguish among the three cases, without even looking at the story context. In addition, combining our stylistic features with language model predictions reaches state of the art performance on the story cloze challenge. Our results demonstrate that different task framings can dramatically affect the way people write.
In this paper, we investigate large-scale zero-shot activity recognition by modeling the visual and linguistic attributes of action verbs. For example, the verb “salute” has several properties, such as being a light movement, a social act, and short in duration. We use these attributes as the internal mapping between visual and textual representations to reason about a previously unseen action. In contrast to much prior work that assumes access to gold standard attributes for zero-shot classes and focuses primarily on object attributes, our model uniquely learns to infer action attributes from dictionary definitions and distributed word representations. Experimental results confirm that action attributes inferred from language can provide a predictive signal for zero-shot prediction of previously unseen activities.
Understanding a long document requires tracking how entities are introduced and evolve over time. We present a new type of language model, EntityNLM, that can explicitly model entities, dynamically update their representations, and contextually generate their mentions. Our model is generative and flexible; it can model an arbitrary number of entities in context while generating each entity mention at an arbitrary length. In addition, it can be used for several different tasks such as language modeling, coreference resolution, and entity prediction. Experimental results with all these tasks demonstrate that our model consistently outperforms strong baselines and prior work.
The framing of an action influences how we perceive its actor. We introduce connotation frames of power and agency, a pragmatic formalism organized using frame semantic representations, to model how different levels of power and agency are implicitly projected on actors through their actions. We use the new power and agency frames to measure the subtle, but prevalent, gender bias in the portrayal of modern film characters and provide insights that deviate from the well-known Bechdel test. Our contributions include an extended lexicon of connotation frames along with a web interface that provides a comprehensive analysis through the lens of connotation frames.
We present an analytic study on the language of news media in the context of political fact-checking and fake news detection. We compare the language of real news with that of satire, hoaxes, and propaganda to find linguistic characteristics of untrustworthy text. To probe the feasibility of automatic political fact-checking, we also present a case study based on PolitiFact.com using their factuality judgments on a 6-point scale. Experiments show that while media fact-checking remains to be an open research question, stylistic cues can help determine the truthfulness of text.
We present a new tree based approach to composing expressive image descriptions that makes use of naturally occuring web images with captions. We investigate two related tasks: image caption generalization and generation, where the former is an optional subtask of the latter. The high-level idea of our approach is to harvest expressive phrases (as tree fragments) from existing image descriptions, then to compose a new description by selectively combining the extracted (and optionally pruned) tree fragments. Key algorithmic components are tree composition and compression, both integrating tree structure with sequence structure. Our proposed system attains significantly better performance than previous approaches for both image caption generalization and generation. In addition, our work is the first to show the empirical benefit of automatically generalized captions for composing natural image descriptions.