This work explores the feasibility of eliciting knowledge from language models (LMs) to decode symbolism, recognizing something (e.g.,roses) as a stand-in for another (e.g., love). We present our evaluative framework, Symbolism Analysis (SymbA), which compares LMs (e.g., RoBERTa, GPT-J) on different types of symbolism and analyze the outcomes along multiple metrics. Our findings suggest that conventional symbols are more reliably elicited from LMs while situated symbols are more challenging. Results also reveal the negative impact of the bias in pre-trained corpora. We further demonstrate that a simple re-ranking strategy can mitigate the bias and significantly improve model performances to be on par with human performances in some cases.

Certain types of classification problems may be performed at multiple levels of granularity; for example, we might want to know the sentiment polarity of a document or a sentence, or a phrase. Often, the prediction at a greater-context (e.g., sentences or paragraphs) may be informative for a more localized prediction at a smaller semantic unit (e.g., words or phrases). However, directly inferring the most salient local features from the global prediction may overlook the semantics of this relationship. This work argues that inference along the contraposition relationship of the local prediction and the corresponding global prediction makes an inference framework that is more accurate and robust to noise. We show how this contraposition framework can be implemented as a transfer function that rewrites a greater-context from one class to another and demonstrate how an appropriate transfer function can be trained from a noisy user-generated corpus. The experimental results validate our insight that the proposed contrapositive framework outperforms the alternative approaches on resource-constrained problem domains.

We investigate the impact of political ideology biases in training data. Through a set of comparison studies, we examine the propagation of biases in several widely-used NLP models and its effect on the overall retrieval accuracy. Our work highlights the susceptibility of large, complex models to propagating the biases from human-selected input, which may lead to a deterioration of retrieval accuracy, and the importance of controlling for these biases. Finally, as a way to mitigate the bias, we propose to learn a text representation that is invariant to political ideology while still judging topic relevance.

Data augmentation has been shown to be effective in providing more training data for machine learning and resulting in more robust classifiers. However, for some problems, there may be multiple augmentation heuristics, and the choices of which one to use may significantly impact the success of the training. In this work, we propose a metric for evaluating augmentation heuristics; specifically, we quantify the extent to which an example is “hard to distinguish” by considering the difference between the distribution of the augmented samples of different classes. Experimenting with multiple heuristics in two prediction tasks (positive/negative sentiment and verbosity/conciseness) validates our claims by revealing the connection between the distribution difference of different classes and the classification accuracy.

Pleonasms are words that are redundant. To aid the development of systems that detect pleonasms in text, we introduce an annotated corpus of semantic pleonasms. We validate the integrity of the corpus with interannotator agreement analyses. We also compare it against alternative resources in terms of their effects on several automatic redundancy detection methods.

Many idiomatic expressions can be interpreted figuratively or literally depending on their contexts. This paper proposes an unsupervised learning method for recognizing the intended usages of idioms. We treat the usages as a latent variable in probabilistic models and train them in a linguistically motivated feature space. Crucially, we show that distributional semantics is a helpful heuristic for distinguishing the literal usage of idioms, giving us a way to formulate a literal usage metric to estimate the likelihood that the idiom is intended literally. This information then serves as a form of distant supervision to guide the unsupervised training process for the probabilistic models. Experiments show that our overall model performs competitively against supervised methods.

This paper presents ArgRewrite, a corpus of between-draft revisions of argumentative essays. Drafts are manually aligned at the sentence level, and the writer’s purpose for each revision is annotated with categories analogous to those used in argument mining and discourse analysis. The corpus should enable advanced research in writing comparison and revision analysis, as demonstrated via our own studies of student revision behavior and of automatic revision purpose prediction.

Generating fluent and grammatical sentences is a major goal for both Machine Translation (MT) and second-language Grammar Error Correction (GEC), but there have not been a lot of cross-fertilization between the two research communities. Arguably, an automatic translate-to-English system might be seen as an English as a Second Language (ESL) writer whose native language is the source language. This paper investigates whether research findings from the GEC community may help with characterizing MT error analysis. We describe a method for the automatic classification of MT errors according to English as a Second Language (ESL) error categories and conduct a large comparison experiment that includes both high-performing and low-performing translate-to-English MT systems for several source languages. Comparing the distribution of MT error types for all the systems suggests that MT systems have fairly similar distributions regardless of their source languages, and the high-performing MT systems have error distributions that are more similar to those of the low-performing MT systems than to those of ESL learners with the same L1.

This paper investigates varying the decoder weight of the language model (LM) when translating different parts of a sentence. We determine the condition under which the LM weight should be adapted. We find that a better translation can be achieved by varying the LM weight when decoding the most problematic spot in a sentence, which we refer to as a difficult segment. Two adaptation strategies are proposed and compared through experiments. We find that adapting a different LM weight for every difficult segment resulted in the largest improvement in translation quality.

Lexical mappings (word translations) between languages are an invaluable resource for multilingual processing. While the problem of extracting lexical mappings from parallel corpora is well-studied, the task is more challenging when the language samples are from non-parallel corpora. The goal of this work is to investigate one such scenario: finding lexical mappings between dialects of a diglossic language, in which people conduct their written communications in a prestigious formal dialect, but they communicate verbally in a colloquial dialect. Because the two dialects serve different socio-linguistic functions, parallel corpora do not naturally exist between them. An example of a diglossic dialect pair is Modern Standard Arabic (MSA) and Levantine Arabic. In this paper, we evaluate the applicability of a standard algorithm for inducing lexical mappings between comparable corpora (Rapp, 1999) to such diglossic corpora pairs. The focus of the paper is an in-depth error analysis, exploring the notion of relatedness in diglossic corpora and scrutinizing the effects of various dimensions of relatedness (such as mode, topic, style, and statistics) on the quality of the resulting translation lexicon.

The frequent occurrence of divergenceS—structural differences between languages—presents a great challenge for statistical word-level alignment. In this paper, we introduce DUSTer, a method for systematically identifying common divergence types and transforming an English sentence structure to bear a closer resemblance to that of another language. Our ultimate goal is to enable more accurate alignment and projection of dependency trees in another language without requiring any training on dependency-tree data in that language. We present an empirical analysis comparing the complexities of performing word-level alignments with and without divergence handling. Our results suggest that our approach facilitates word-level alignment, particularly for sentence pairs containing divergences.