Much recent work seeks to evaluate values and opinions in large language models (LLMs) using multiple-choice surveys and questionnaires. Most of this work is motivated by concerns around real-world LLM applications. For example, politically-biased LLMs may subtly influence society when they are used by millions of people. Such real-world concerns, however, stand in stark contrast to the artificiality of current evaluations: real users do not typically ask LLMs survey questions. Motivated by this discrepancy, we challenge the prevailing *constrained* evaluation paradigm for values and opinions in LLMs and explore more realistic *unconstrained* evaluations. As a case study, we focus on the popular Political Compass Test (PCT). In a systematic review, we find that most prior work using the PCT *forces models to comply with the PCT’s multiple-choice format. We show that models give substantively different answers when not forced; that answers change depending on how models are forced; and that answers lack paraphrase robustness. Then, we demonstrate that models give different answers yet again in a more realistic open-ended answer setting. We distill these findings into recommendations and open challenges in evaluating values and opinions in LLMs.

Language models (LMs) have become ubiquitous in both NLP research and in commercial product offerings. As their commercial importance has surged, the most powerful models have become closed off, gated behind proprietary interfaces, with important details of their training data, architectures, and development undisclosed. Given the importance of these details in scientifically studying these models, including their biases and potential risks, we believe it is essential for the research community to have access to powerful, truly open LMs. To this end, we have built OLMo, a competitive, truly Open Language Model, to enable the scientific study of language models. Unlike most prior efforts that have only released model weights and inference code, we release OLMo alongside open training data and training and evaluation code. We hope this release will empower the open research community and inspire a new wave of innovation.

Detecting semantic arguments of a predicate word has been conventionally modeled as a sentence-level task. The typical reader, however, perfectly interprets predicate-argument relations in a much wider context than just the sentence where the predicate was evoked. In this work, we reformulate the problem of argument detection through textual entailment to capture semantic relations across sentence boundaries. We propose a method that tests whether some semantic relation can be inferred from a full passage by first encoding it into a simple and standalone proposition and then testing for entailment against the passage. Our method does not require direct supervision, which is generally absent due to dataset scarcity, but instead builds on existing NLI and sentence-level SRL resources. Such a method can potentially explicate pragmatically understood relations into a set of explicit sentences. We demonstrate it on a recent document-level benchmark, outperforming some supervised methods and contemporary language models.

Prompt-based methods have been used extensively across NLP to build zero- and few-shot label predictors. Many NLP tasks are naturally structured: that is, their outputs consist of multiple labels which constrain each other. Annotating data for such tasks can be cumbersome. Can the promise of the prompt-based paradigm be extended to such structured outputs? In this paper, we present a framework for constructing zero- and few-shot linguistic structure predictors. Our key insight is that we can use structural constraints—and combinatorial inference derived from them—to filter out inconsistent structures predicted by large language models. We instantiated this framework on two structured prediction tasks, and five datasets. Across all cases, our results show that enforcing consistency not only constructs structurally valid outputs, but also improves performance over the unconstrained variants.

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.

Tasks involving text generation based on multiple input texts, such as multi-document summarization, long-form question answering and contemporary dialogue applications, challenge models for their ability to properly consolidate partly-overlapping multi-text information. However, these tasks entangle the consolidation phase with the often subjective and ill-defined content selection requirement, impeding proper assessment of models’ consolidation capabilities. In this paper, we suggest revisiting the sentence union generation task as an effective well-defined testbed for assessing text consolidation capabilities, decoupling the consolidation challenge from subjective content selection. To support research on this task, we present refined annotation methodology and tools for crowdsourcing sentence union, create the largest union dataset to date and provide an analysis of its rich coverage of various consolidation aspects. We then propose a comprehensive evaluation protocol for union generation, including both human and automatic evaluation. Finally, as baselines, we evaluate state-of-the-art language models on the task, along with a detailed analysis of their capacity to address multi-text consolidation challenges and their limitations.

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.

Disagreement in natural language annotation has mostly been studied from a perspective of biases introduced by the annotators and the annotation frameworks. Here, we propose to analyze another source of bias—task design bias, which has a particularly strong impact on crowdsourced linguistic annotations where natural language is used to elicit the interpretation of lay annotators. For this purpose we look at implicit discourse relation annotation, a task that has repeatedly been shown to be difficult due to the relations’ ambiguity. We compare the annotations of 1,200 discourse relations obtained using two distinct annotation tasks and quantify the biases of both methods across four different domains. Both methods are natural language annotation tasks designed for crowdsourcing. We show that the task design can push annotators towards certain relations and that some discourse relation senses can be better elicited with one or the other annotation approach. We also conclude that this type of bias should be taken into account when training and testing models.

Various works suggest the appeals of incorporating explicit semantic representations when addressing challenging realistic NLP scenarios. Common approaches offer either comprehensive linguistically-based formalisms, like AMR, or alternatively Open-IE, which provides a shallow and partial representation. More recently, an appealing trend introduces semi-structured natural-language structures as an intermediate meaning-capturing representation, often in the form of questions and answers.In this work, we further promote this line of research by considering three prior QA-based semantic representations. These cover verbal, nominalized and discourse-based predications, regarded as jointly providing a comprehensive representation of textual information — termed QASem. To facilitate this perspective, we investigate how to best utilize pre-trained sequence-to-sequence language models, which seem particularly promising for generating representations that consist of natural language expressions (questions and answers). In particular, we examine and analyze input and output linearization strategies, as well as data augmentation and multitask learning for a scarce training data setup. Consequently, we release the first unified QASem parsing tool, easily applicable for downstream tasks that can benefit from an explicit semi-structured account of information units in text.

Figurative language (e.g., “he flew like the wind”) is challenging to understand, as it is hard to tell what implicit information is being conveyed from the surface form alone. We hypothesize that to perform this task well, the reader needs to mentally elaborate the scene being described to identify a sensible meaning of the language. We present DREAM-FLUTE, a figurative language understanding system that does this, first forming a “mental model” of situations described in a premise and hypothesis before making an entailment/contradiction decision and generating an explanation. DREAM-FLUTE uses an existing scene elaboration model, DREAM, for constructing its “mental model.” In the FigLang2022 Shared Task evaluation, DREAM-FLUTE achieved (joint) first place (Acc@60=63.3%), and can perform even better with ensemble techniques, demonstrating the effectiveness of this approach. More generally, this work suggests that adding a reflective component to pretrained language models can improve their performance beyond standard fine-tuning (3.3% improvement in Acc@60).

Obtaining linguistic annotation from novice crowdworkers is far from trivial. A case in point is the annotation of discourse relations, which is a complicated task. Recent methods have obtained promising results by extracting relation labels from either discourse connectives (DCs) or question-answer (QA) pairs that participants provide. The current contribution studies the effect of worker selection and training on the agreement on implicit relation labels between workers and gold labels, for both the DC and the QA method. In Study 1, workers were not specifically selected or trained, and the results show that there is much room for improvement. Study 2 shows that a combination of selection and training does lead to improved results, but the method is cost- and time-intensive. Study 3 shows that a selection-only approach is a viable alternative; it results in annotations of comparable quality compared to annotations from trained participants. The results generalized over both the DC and QA method and therefore indicate that a selection-only approach could also be effective for other crowdsourced discourse annotation tasks.

Abstraction is a core tenet of human cognition and communication. When composing natural language instructions, humans naturally evoke abstraction to convey complex procedures in an efficient and concise way. Yet, interpreting and grounding abstraction expressed in NL has not yet been systematically studied in NLP, with no accepted benchmarks specifically eliciting abstraction in NL. In this work, we set the foundation for a systematic study of processing and grounding abstraction in NLP. First, we deliver a novel abstraction elicitation method and present Hexagons, a 2D instruction-following game. Using Hexagons we collected over 4k naturally occurring visually-grounded instructions rich with diverse types of abstractions. From these data, we derive an instruction-to-execution task and assess different types of neural models. Our results show that contemporary models and modeling practices are substantially inferior to human performance, and that model performance is inversely correlated with the level of abstraction, showing less satisfying performance on higher levels of abstraction. These findings are consistent across models and setups, confirming that abstraction is a challenging phenomenon deserving further attention and study in NLP/AI research.

Modality is the linguistic ability to describe vents with added information such as how desirable, plausible, or feasible they are. Modality is important for many NLP downstream tasks such as the detection of hedging, uncertainty, speculation, and more. Previous studies that address modality detection in NLP often restrict modal expressions to a closed syntactic class, and the modal sense labels are vastly different across different studies, lacking an accepted standard. Furthermore, these senses are often analyzed independently of the events that they modify. This work builds on the theoretical foundations of the Georgetown Gradable Modal Expressions (GME) work by Rubinstein et al. (2013) to propose an event-based modality detection task where modal expressions can be words of any syntactic class and sense labels are drawn from a comprehensive taxonomy which harmonizes the modal concepts contributed by the different studies. We present experiments on the GME corpus aiming to detect and classify fine-grained modal concepts and associate them with their modified events. We show that detecting and classifying modal expressions is not only feasible, it also improves the detection of modal events in their own right.

Asking questions about a situation is an inherent step towards understanding it. To this end, we introduce the task of role question generation, which, given a predicate mention and a passage, requires producing a set of questions asking about all possible semantic roles of the predicate. We develop a two-stage model for this task, which first produces a context-independent question prototype for each role and then revises it to be contextually appropriate for the passage. Unlike most existing approaches to question generation, our approach does not require conditioning on existing answers in the text. Instead, we condition on the type of information to inquire about, regardless of whether the answer appears explicitly in the text, could be inferred from it, or should be sought elsewhere. Our evaluation demonstrates that we generate diverse and well-formed questions for a large, broad-coverage ontology of predicates and roles.

We propose a new semantic scheme for capturing predicate-argument relations for nominalizations, termed QANom. This scheme extends the QA-SRL formalism (He et al., 2015), modeling the relations between nominalizations and their arguments via natural language question-answer pairs. We construct the first QANom dataset using controlled crowdsourcing, analyze its quality and compare it to expertly annotated nominal-SRL annotations, as well as to other QA-driven annotations. In addition, we train a baseline QANom parser for identifying nominalizations and labeling their arguments with question-answer pairs. Finally, we demonstrate the extrinsic utility of our annotations for downstream tasks using both indirect supervision and zero-shot settings.

Discourse relations describe how two propositions relate to one another, and identifying them automatically is an integral part of natural language understanding. However, annotating discourse relations typically requires expert annotators. Recently, different semantic aspects of a sentence have been represented and crowd-sourced via question-and-answer (QA) pairs. This paper proposes a novel representation of discourse relations as QA pairs, which in turn allows us to crowd-source wide-coverage data annotated with discourse relations, via an intuitively appealing interface for composing such questions and answers. Based on our proposed representation, we collect a novel and wide-coverage QADiscourse dataset, and present baseline algorithms for predicting QADiscourse relations.

Sense classification of discourse relations is a sub-task of shallow discourse parsing. Discourse relations can occur both across sentences (inter-sentential) and within sentences (intra-sentential), and more than one discourse relation can hold between the same units. Using a newly available corpus of discourse-annotated intra-sentential conjoined verb phrases, we demonstrate a sequential classification pipeline for their multi-label sense classification. We assess the importance of each feature used in the classification, the feature scope, and what is lost in moving from gold standard manual parses to the output of an off-the-shelf parser.