Maarten Sap


2021

pdf bib
Challenges in Automated Debiasing for Toxic Language Detection
Xuhui Zhou | Maarten Sap | Swabha Swayamdipta | Yejin Choi | Noah Smith
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume

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.

pdf bib
Detoxifying Language Models Risks Marginalizing Minority Voices
Albert Xu | Eshaan Pathak | Eric Wallace | Suchin Gururangan | Maarten Sap | Dan Klein
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

Language models (LMs) must be both safe and equitable to be responsibly deployed in practice. With safety in mind, numerous detoxification techniques (e.g., Dathathri et al. 2020; Krause et al. 2020) have been proposed to mitigate toxic LM generations. In this work, we show that these detoxification techniques hurt equity: they decrease the utility of LMs on language used by marginalized groups (e.g., African-American English and minority identity mentions). In particular, we perform automatic and human evaluations of text generation quality when LMs are conditioned on inputs with different dialects and group identifiers. We find that detoxification makes LMs more brittle to distribution shift, especially on language used by marginalized groups. We identify that these failures stem from detoxification methods exploiting spurious correlations in toxicity datasets. Overall, our results highlight the tension between the controllability and distributional robustness of LMs.

pdf bib
DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts
Alisa Liu | Maarten Sap | Ximing Lu | Swabha Swayamdipta | Chandra Bhagavatula | Noah A. Smith | Yejin Choi
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)

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.

pdf bib
Proceedings of the 1st Workshop on NLP for Positive Impact
Anjalie Field | Shrimai Prabhumoye | Maarten Sap | Zhijing Jin | Jieyu Zhao | Chris Brockett
Proceedings of the 1st Workshop on NLP for Positive Impact

2020

pdf bib
RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models
Samuel Gehman | Suchin Gururangan | Maarten Sap | Yejin Choi | Noah A. Smith
Findings of the Association for Computational Linguistics: EMNLP 2020

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.

pdf bib
Recollection versus Imagination: Exploring Human Memory and Cognition via Neural Language Models
Maarten Sap | Eric Horvitz | Yejin Choi | Noah A. Smith | James Pennebaker
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

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.

pdf bib
Social Bias Frames: Reasoning about Social and Power Implications of Language
Maarten Sap | Saadia Gabriel | Lianhui Qin | Dan Jurafsky | Noah A. Smith | Yejin Choi
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

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.

pdf bib
Commonsense Reasoning for Natural Language Processing
Maarten Sap | Vered Shwartz | Antoine Bosselut | Yejin Choi | Dan Roth
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: Tutorial Abstracts

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.

pdf bib
Exploring the Effect of Author and Reader Identity in Online Story Writing: the STORIESINTHEWILD Corpus.
Tal August | Maarten Sap | Elizabeth Clark | Katharina Reinecke | Noah A. Smith
Proceedings of the First Joint Workshop on Narrative Understanding, Storylines, and Events

Current story writing or story editing systems rely on human judgments of story quality for evaluating performance, often ignoring the subjectivity in ratings. We analyze the effect of author and reader characteristics and story writing setup on the quality of stories in a short storytelling task. To study this effect, we create and release STORIESINTHEWILD, containing 1,630 stories collected on a volunteer-based crowdsourcing platform. Each story is rated by three different readers, and comes paired with the author’s and reader’s age, gender, and personality. Our findings show significant effects of authors’ and readers’ identities, as well as writing setup, on story writing and ratings. Notably, compared to younger readers, readers age 45 and older consider stories significantly less creative and less entertaining. Readers also prefer stories written all at once, rather than in chunks, finding them more coherent and creative. We also observe linguistic differences associated with authors’ demographics (e.g., older authors wrote more vivid and emotional stories). Our findings suggest that reader and writer demographics, as well as writing setup, should be accounted for in story writing evaluations.

pdf bib
Social Chemistry 101: Learning to Reason about Social and Moral Norms
Maxwell Forbes | Jena D. Hwang | Vered Shwartz | Maarten Sap | Yejin Choi
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

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.

pdf bib
PowerTransformer: Unsupervised Controllable Revision for Biased Language Correction
Xinyao Ma | Maarten Sap | Hannah Rashkin | Yejin Choi
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

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.

2019

pdf bib
Social IQa: Commonsense Reasoning about Social Interactions
Maarten Sap | Hannah Rashkin | Derek Chen | Ronan Le Bras | Yejin Choi
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)

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).

pdf bib
The Risk of Racial Bias in Hate Speech Detection
Maarten Sap | Dallas Card | Saadia Gabriel | Yejin Choi | Noah A. Smith
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

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.

pdf bib
COMET: Commonsense Transformers for Automatic Knowledge Graph Construction
Antoine Bosselut | Hannah Rashkin | Maarten Sap | Chaitanya Malaviya | Asli Celikyilmaz | Yejin Choi
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

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.

2018

pdf bib
Sounding Board: A User-Centric and Content-Driven Social Chatbot
Hao Fang | Hao Cheng | Maarten Sap | Elizabeth Clark | Ari Holtzman | Yejin Choi | Noah A. Smith | Mari Ostendorf
Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations

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.

pdf bib
Event2Mind: Commonsense Inference on Events, Intents, and Reactions
Hannah Rashkin | Maarten Sap | Emily Allaway | Noah A. Smith | Yejin Choi
Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

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.

pdf bib
Modeling Naive Psychology of Characters in Simple Commonsense Stories
Hannah Rashkin | Antoine Bosselut | Maarten Sap | Kevin Knight | Yejin Choi
Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

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.

2017

pdf bib
Story Cloze Task: UW NLP System
Roy Schwartz | Maarten Sap | Ioannis Konstas | Leila Zilles | Yejin Choi | Noah A. Smith
Proceedings of the 2nd Workshop on Linking Models of Lexical, Sentential and Discourse-level Semantics

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).

pdf bib
The Effect of Different Writing Tasks on Linguistic Style: A Case Study of the ROC Story Cloze Task
Roy Schwartz | Maarten Sap | Ioannis Konstas | Leila Zilles | Yejin Choi | Noah A. Smith
Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL 2017)

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.

pdf bib
Connotation Frames of Power and Agency in Modern Films
Maarten Sap | Marcella Cindy Prasettio | Ari Holtzman | Hannah Rashkin | Yejin Choi
Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

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.

pdf bib
DLATK: Differential Language Analysis ToolKit
H. Andrew Schwartz | Salvatore Giorgi | Maarten Sap | Patrick Crutchley | Lyle Ungar | Johannes Eichstaedt
Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing: System Demonstrations

We present Differential Language Analysis Toolkit (DLATK), an open-source python package and command-line tool developed for conducting social-scientific language analyses. While DLATK provides standard NLP pipeline steps such as tokenization or SVM-classification, its novel strengths lie in analyses useful for psychological, health, and social science: (1) incorporation of extra-linguistic structured information, (2) specified levels and units of analysis (e.g. document, user, community), (3) statistical metrics for continuous outcomes, and (4) robust, proven, and accurate pipelines for social-scientific prediction problems. DLATK integrates multiple popular packages (SKLearn, Mallet), enables interactive usage (Jupyter Notebooks), and generally follows object oriented principles to make it easy to tie in additional libraries or storage technologies.

2015

pdf bib
The role of personality, age, and gender in tweeting about mental illness
Daniel Preoţiuc-Pietro | Johannes Eichstaedt | Gregory Park | Maarten Sap | Laura Smith | Victoria Tobolsky | H. Andrew Schwartz | Lyle Ungar
Proceedings of the 2nd Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality

pdf bib
Mental Illness Detection at the World Well-Being Project for the CLPsych 2015 Shared Task
Daniel Preoţiuc-Pietro | Maarten Sap | H. Andrew Schwartz | Lyle Ungar
Proceedings of the 2nd Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality

pdf bib
Extracting Human Temporal Orientation from Facebook Language
H. Andrew Schwartz | Gregory Park | Maarten Sap | Evan Weingarten | Johannes Eichstaedt | Margaret Kern | David Stillwell | Michal Kosinski | Jonah Berger | Martin Seligman | Lyle Ungar
Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

2014

pdf bib
Towards Assessing Changes in Degree of Depression through Facebook
H. Andrew Schwartz | Johannes Eichstaedt | Margaret L. Kern | Gregory Park | Maarten Sap | David Stillwell | Michal Kosinski | Lyle Ungar
Proceedings of the Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality

pdf bib
Developing Age and Gender Predictive Lexica over Social Media
Maarten Sap | Gregory Park | Johannes Eichstaedt | Margaret Kern | David Stillwell | Michal Kosinski | Lyle Ungar | Hansen Andrew Schwartz
Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)