Anjalie Field

Anjalie Field, Prateek Verma, Nay San et al. · stanford

Police body-worn cameras have the potential to improve accountability and transparency in policing. Yet in practice, they result in millions of hours of footage that is never reviewed. We investigate the potential of large pre-trained speech models for facilitating reviews, focusing on ASR and officer speech detection in footage from traffic stops. Our proposed pipeline includes training data alignment and filtering, fine-tuning with resource constraints, and combining officer speech detection with ASR for a fully automated approach. We find that (1) fine-tuning strongly improves ASR performance on officer speech (WER=12-13%), (2) ASR on officer speech is much more accurate than on community member speech (WER=43.55-49.07%), (3) domain-specific tasks like officer speech detection and diarization remain challenging. Our work offers practical applications for reviewing body camera footage and general guidance for adapting pre-trained speech models to noisy multi-speaker domains.

Chan Young Park, Julia Mendelsohn, Anjalie Field et al. · cmu

NLP research on public opinion manipulation campaigns has primarily focused on detecting overt strategies such as fake news and disinformation. However, information manipulation in the ongoing Russia-Ukraine war exemplifies how governments and media also employ more nuanced strategies. We release a new dataset, VoynaSlov, containing 38M+ posts from Russian media outlets on Twitter and VKontakte, as well as public activity and responses, immediately preceding and during the 2022 Russia-Ukraine war. We apply standard and recently-developed NLP models on VoynaSlov to examine agenda setting, framing, and priming, several strategies underlying information manipulation, and reveal variation across media outlet control, social media platform, and time. Our examination of these media effects and extensive discussion of current approaches' limitations encourage further development of NLP models for understanding information manipulation in emerging crises, as well as other real-world and interdisciplinary tasks.

Inna Wanyin Lin, Lucille Njoo, Anjalie Field et al. · uw

Mental health stigma prevents many individuals from receiving the appropriate care, and social psychology studies have shown that mental health tends to be overlooked in men. In this work, we investigate gendered mental health stigma in masked language models. In doing so, we operationalize mental health stigma by developing a framework grounded in psychology research: we use clinical psychology literature to curate prompts, then evaluate the models' propensity to generate gendered words. We find that masked language models capture societal stigma about gender in mental health: models are consistently more likely to predict female subjects than male in sentences about having a mental health condition (32% vs. 19%), and this disparity is exacerbated for sentences that indicate treatment-seeking behavior. Furthermore, we find that different models capture dimensions of stigma differently for men and women, associating stereotypes like anger, blame, and pity more with women with mental health conditions than with men. In showing the complex nuances of models' gendered mental health stigma, we demonstrate that context and overlapping dimensions of identity are important considerations when assessing computational models' social biases.

Nupoor Gandhi, Anjalie Field, Emma Strubell

Although recent neural models for coreference resolution have led to substantial improvements on benchmark datasets, transferring these models to new target domains containing out-of-vocabulary spans and requiring differing annotation schemes remains challenging. Typical approaches involve continued training on annotated target-domain data, but obtaining annotations is costly and time-consuming. We show that annotating mentions alone is nearly twice as fast as annotating full coreference chains. Accordingly, we propose a method for efficiently adapting coreference models, which includes a high-precision mention detection objective and requires annotating only mentions in the target domain. Extensive evaluation across three English coreference datasets: CoNLL-2012 (news/conversation), i2b2/VA (medical notes), and previously unstudied child welfare notes, reveals that our approach facilitates annotation-efficient transfer and results in a 7-14% improvement in average F1 without increasing annotator time.

Yaxuan Ren, Krithika Ramesh, Yaxing Yao et al.

In this work, we aim to clarify and reconcile metrics for evaluating privacy protection in text through a systematic survey. Although text anonymization is essential for enabling NLP research and model development in domains with sensitive data, evaluating whether anonymization methods sufficiently protect privacy remains an open challenge. In manually reviewing 47 papers that report privacy metrics, we identify and compare six distinct privacy notions, and analyze how the associated metrics capture different aspects of privacy risk. We then assess how well these notions align with legal privacy standards (HIPAA and GDPR), as well as user-centered expectations grounded in HCI studies. Our analysis offers practical guidance on navigating the landscape of privacy evaluation approaches further and highlights gaps in current practices. Ultimately, we aim to facilitate more robust, comparable, and legally aware privacy evaluations in text anonymization.

Jonathan Ivey, Anjalie Field, Ziang Xiao

Qualitative interviews provide essential insights into human experiences when they elicit high-quality responses. While qualitative and NLP researchers have proposed various measures of interview quality, these measures lack validation that high-scoring responses actually contribute to the study's goals. In this work, we identify, implement, and evaluate 10 proposed measures of interview response quality to determine which are actually predictive of a response's contribution to the study findings. To conduct our analysis, we introduce the Qualitative Interview Corpus, a newly constructed dataset of 343 interview transcripts with 16,940 participant responses from 14 real research projects. We find that direct relevance to a key research question is the strongest predictor of response quality. We additionally find that two measures commonly used to evaluate NLP interview systems, clarity and surprisal-based informativeness, are not predictive of response quality. Our work provides analytic insights and grounded, scalable metrics to inform the design of qualitative studies and the evaluation of automated interview systems.

Sandesh S Rangreji, Mian Zhong, Anjalie Field

Analyses of document collections often require selecting what data to analyze, as not all documents are relevant to a particular research question and computational constraints preclude analyzing all documents, yet little work has examined effects of selection strategy choices. We systematically evaluate seven selection methods (from random selection to hybrid retrieval) on outputs from four text analyses methods (LDA, BERTopic, TopicGPT, HiCode) over two datasets with 26 open-ended queries. Our evaluation reveals practice guidance: semantic or hybrid retrieval offer strong go-to approaches that avoid the pitfalls of weaker selection strategies and the unnecessary compute overhead of more complicated ones. Overall, our evaluation framework establishes data selection as a methodological decision, rather than a practical necessity, inviting the development of new strategies.

Taryn Wong, Zeerak Talat, Hanan Aldarmaki et al.

Critical analyses of emotion recognition technology have raised ethical concerns around task validity and potential downstream impacts, urging researchers to ensure alignment between their stated motivations and practice. However, these discussions have not adequately influenced or drawn from research on speech emotion recognition (SER). We address this gap by conducting a systematic survey of SER research to uncover what stated motivations drive this work and if they align with the datasets and emotions studied. We find that while SER research identifies appealing goals, such as well-situated voice-activated systems or healthcare applications, commonly-used datasets do not reflect these proposed deployment contexts, thus presenting a gap between motivations and research practices. We argue that such gaps engender ethical concerns, and that SER research should reassert itself with concrete use-cases to prevent misinterpretations, misuse, and downstream harms.

Maria Antoniak, Anjalie Field, Jimin Mun et al. · allen-ai, cmu

Riveter provides a complete easy-to-use pipeline for analyzing verb connotations associated with entities in text corpora. We prepopulate the package with connotation frames of sentiment, power, and agency, which have demonstrated usefulness for capturing social phenomena, such as gender bias, in a broad range of corpora. For decades, lexical frameworks have been foundational tools in computational social science, digital humanities, and natural language processing, facilitating multifaceted analysis of text corpora. But working with verb-centric lexica specifically requires natural language processing skills, reducing their accessibility to other researchers. By organizing the language processing pipeline, providing complete lexicon scores and visualizations for all entities in a corpus, and providing functionality for users to target specific research questions, Riveter greatly improves the accessibility of verb lexica and can facilitate a broad range of future research.

Mian Zhong, Pristina Wang, Anjalie Field

Despite numerous applications for fine-grained corpus analysis, researchers continue to rely on manual labeling, which does not scale, or statistical tools like topic modeling, which are difficult to control. We propose that LLMs have the potential to scale the nuanced analyses that researchers typically conduct manually to large text corpora. To this effect, inspired by qualitative research methods, we develop HICode, a two-part pipeline that first inductively generates labels directly from analysis data and then hierarchically clusters them to surface emergent themes. We validate this approach across three diverse datasets by measuring alignment with human-constructed themes and demonstrating its robustness through automated and human evaluations. Finally, we conduct a case study of litigation documents related to the ongoing opioid crisis in the U.S., revealing aggressive marketing strategies employed by pharmaceutical companies and demonstrating HICode's potential for facilitating nuanced analyses in large-scale data.

Alina Hyk, Kiera McCormick, Mian Zhong et al.

There is growing interest in leveraging LLMs to aid in astronomy and other scientific research, but benchmarks for LLM evaluation in general have not kept pace with the increasingly diverse ways that real people evaluate and use these models. In this study, we seek to improve evaluation procedures by building an understanding of how users evaluate LLMs. We focus on a particular use case: an LLM-powered retrieval-augmented generation bot for engaging with astronomical literature, which we deployed via Slack. Our inductive coding of 368 queries to the bot over four weeks and our follow-up interviews with 11 astronomers reveal how humans evaluated this system, including the types of questions asked and the criteria for judging responses. We synthesize our findings into concrete recommendations for building better benchmarks, which we then employ in constructing a sample benchmark for evaluating LLMs for astronomy. Overall, our work offers ways to improve LLM evaluation and ultimately usability, particularly for use in scientific research.

Krithika Ramesh, Daniel Smolyak, Zihao Zhao et al.

We present SynthTextEval, a toolkit for conducting comprehensive evaluations of synthetic text. The fluency of large language model (LLM) outputs has made synthetic text potentially viable for numerous applications, such as reducing the risks of privacy violations in the development and deployment of AI systems in high-stakes domains. Realizing this potential, however, requires principled consistent evaluations of synthetic data across multiple dimensions: its utility in downstream systems, the fairness of these systems, the risk of privacy leakage, general distributional differences from the source text, and qualitative feedback from domain experts. SynthTextEval allows users to conduct evaluations along all of these dimensions over synthetic data that they upload or generate using the toolkit's generation module. While our toolkit can be run over any data, we highlight its functionality and effectiveness over datasets from two high-stakes domains: healthcare and law. By consolidating and standardizing evaluation metrics, we aim to improve the viability of synthetic text, and in-turn, privacy-preservation in AI development.

Miriam Wanner, Sophia Hager, Anjalie Field

Local news stations are often considered to be reliable sources of non-politicized information, particularly local concerns that residents care about. Because these stations are trusted news sources, viewers are particularly susceptible to the information they report. The Sinclair Broadcast group is a broadcasting company that has acquired many local news stations in the last decade. We investigate the effects of local news stations being acquired by Sinclair: how does coverage change? We use computational methods to investigate changes in internet content put out by local news stations before and after being acquired by Sinclair and in comparison to national news outlets. We find that there is clear evidence that local news stations report more frequently on national news at the expense of local topics, and that their coverage of polarizing national topics increases.

Zihao Zhao, Anjalie Field

Text anonymization is essential for responsibly developing and deploying AI in high-stakes domains such as healthcare, social services, and law. In this work, we propose a novel methodology for privacy-preserving synthetic text generation that leverages the principles of de-identification and the Hiding In Plain Sight (HIPS) theory. Our approach introduces entity-aware control codes to guide controllable generation using either in-context learning (ICL) or prefix tuning. The ICL variant ensures privacy levels consistent with the underlying de-identification system, while the prefix tuning variant incorporates a custom masking strategy and loss function to support scalable, high-quality generation. Experiments on legal and clinical datasets demonstrate that our method achieves a strong balance between privacy protection and utility, offering a practical and effective solution for synthetic text generation in sensitive domains.

Anjalie Field, Amanda Coston, Nupoor Gandhi et al.

Although much literature has established the presence of demographic bias in natural language processing (NLP) models, most work relies on curated bias metrics that may not be reflective of real-world applications. At the same time, practitioners are increasingly using algorithmic tools in high-stakes settings, with particular recent interest in NLP. In this work, we focus on one such setting: child protective services (CPS). CPS workers often write copious free-form text notes about families they are working with, and CPS agencies are actively seeking to deploy NLP models to leverage these data. Given well-established racial bias in this setting, we investigate possible ways deployed NLP is liable to increase racial disparities. We specifically examine word statistics within notes and algorithmic fairness in risk prediction, coreference resolution, and named entity recognition (NER). We document consistent algorithmic unfairness in NER models, possible algorithmic unfairness in coreference resolution models, and little evidence of exacerbated racial bias in risk prediction. While there is existing pronounced criticism of risk prediction, our results expose previously undocumented risks of racial bias in realistic information extraction systems, highlighting potential concerns in deploying them, even though they may appear more benign. Our work serves as a rare realistic examination of NLP algorithmic fairness in a potential deployed setting and a timely investigation of a specific risk associated with deploying NLP in CPS settings.

Nupoor Gandhi, Anjalie Field, Yulia Tsvetkov

Recent work has shown fine-tuning neural coreference models can produce strong performance when adapting to different domains. However, at the same time, this can require a large amount of annotated target examples. In this work, we focus on supervised domain adaptation for clinical notes, proposing the use of concept knowledge to more efficiently adapt coreference models to a new domain. We develop methods to improve the span representations via (1) a retrofitting loss to incentivize span representations to satisfy a knowledge-based distance function and (2) a scaffolding loss to guide the recovery of knowledge from the span representation. By integrating these losses, our model is able to improve our baseline precision and F-1 score. In particular, we show that incorporating knowledge with end-to-end coreference models results in better performance on the most challenging, domain-specific spans.

Anjalie Field, Su Lin Blodgett, Zeerak Waseem et al.

Despite inextricable ties between race and language, little work has considered race in NLP research and development. In this work, we survey 79 papers from the ACL anthology that mention race. These papers reveal various types of race-related bias in all stages of NLP model development, highlighting the need for proactive consideration of how NLP systems can uphold racial hierarchies. However, persistent gaps in research on race and NLP remain: race has been siloed as a niche topic and remains ignored in many NLP tasks; most work operationalizes race as a fixed single-dimensional variable with a ground-truth label, which risks reinforcing differences produced by historical racism; and the voices of historically marginalized people are nearly absent in NLP literature. By identifying where and how NLP literature has and has not considered race, especially in comparison to related fields, our work calls for inclusion and racial justice in NLP research practices.

Anjalie Field, Chan Young Park, Kevin Z. Lin et al.

Social biases on Wikipedia, a widely-read global platform, could greatly influence public opinion. While prior research has examined man/woman gender bias in biography articles, possible influences of other demographic attributes limit conclusions. In this work, we present a methodology for analyzing Wikipedia pages about people that isolates dimensions of interest (e.g., gender), from other attributes (e.g., occupation). Given a target corpus for analysis (e.g.~biographies about women), we present a method for constructing a comparison corpus that matches the target corpus in as many attributes as possible, except the target one. We develop evaluation metrics to measure how well the comparison corpus aligns with the target corpus and then examine how articles about gender and racial minorities (cis. women, non-binary people, transgender women, and transgender men; African American, Asian American, and Hispanic/Latinx American people) differ from other articles. In addition to identifying suspect social biases, our results show that failing to control for covariates can result in different conclusions and veil biases. Our contributions include methodology that facilitates further analyses of bias in Wikipedia articles, findings that can aid Wikipedia editors in reducing biases, and a framework and evaluation metrics to guide future work in this area.

Chan Young Park, Xinru Yan, Anjalie Field et al.

Specific lexical choices in narrative text reflect both the writer's attitudes towards people in the narrative and influence the audience's reactions. Prior work has examined descriptions of people in English using contextual affective analysis, a natural language processing (NLP) technique that seeks to analyze how people are portrayed along dimensions of power, agency, and sentiment. Our work presents an extension of this methodology to multilingual settings, which is enabled by a new corpus that we collect and a new multilingual model. We additionally show how word connotations differ across languages and cultures, highlighting the difficulty of generalizing existing English datasets and methods. We then demonstrate the usefulness of our method by analyzing Wikipedia biography pages of members of the LGBT community across three languages: English, Russian, and Spanish. Our results show systematic differences in how the LGBT community is portrayed across languages, surfacing cultural differences in narratives and signs of social biases. Practically, this model can be used to identify Wikipedia articles for further manual analysis -- articles that might contain content gaps or an imbalanced representation of particular social groups.

Mengzhou Xia, Anjalie Field, Yulia Tsvetkov

In current hate speech datasets, there exists a high correlation between annotators' perceptions of toxicity and signals of African American English (AAE). This bias in annotated training data and the tendency of machine learning models to amplify it cause AAE text to often be mislabeled as abusive/offensive/hate speech with a high false positive rate by current hate speech classifiers. In this paper, we use adversarial training to mitigate this bias, introducing a hate speech classifier that learns to detect toxic sentences while demoting confounds corresponding to AAE texts. Experimental results on a hate speech dataset and an AAE dataset suggest that our method is able to substantially reduce the false positive rate for AAE text while only minimally affecting the performance of hate speech classification.

Anjalie Field, Sascha Rothe, Simon Baumgartner et al.

We evaluate the performance of transformer encoders with various decoders for information organization through a new task: generation of section headings for Wikipedia articles. Our analysis shows that decoders containing attention mechanisms over the encoder output achieve high-scoring results by generating extractive text. In contrast, a decoder without attention better facilitates semantic encoding and can be used to generate section embeddings. We additionally introduce a new loss function, which further encourages the decoder to generate high-quality embeddings.

Aman Tyagi, Anjalie Field, Priyank Lathwal et al.

Between February 14, 2019 and March 4, 2019, a terrorist attack in Pulwama, Kashmir followed by retaliatory airstrikes led to rising tensions between India and Pakistan, two nuclear-armed countries. In this work, we examine polarizing messaging on Twitter during these events, particularly focusing on the positions of Indian and Pakistani politicians. We use a label propagation technique focused on hashtag co-occurrences to find polarizing tweets and users. Our analysis reveals that politicians in the ruling political party in India (BJP) used polarized hashtags and called for escalation of conflict more so than politicians from other parties. Our work offers the first analysis of how escalating tensions between India and Pakistan manifest on Twitter and provides a framework for studying polarizing messages.

Anjalie Field, Yulia Tsvetkov

Despite their prevalence in society, social biases are difficult to identify, primarily because human judgements in this domain can be unreliable. We take an unsupervised approach to identifying gender bias against women at a comment level and present a model that can surface text likely to contain bias. Our main challenge is forcing the model to focus on signs of implicit bias, rather than other artifacts in the data. Thus, our methodology involves reducing the influence of confounds through propensity matching and adversarial learning. Our analysis shows how biased comments directed towards female politicians contain mixed criticisms, while comments directed towards other female public figures focus on appearance and sexualization. Ultimately, our work offers a way to capture subtle biases in various domains without relying on subjective human judgements.

Anjalie Field, Yulia Tsvetkov

While contextualized word representations have improved state-of-the-art benchmarks in many NLP tasks, their potential usefulness for social-oriented tasks remains largely unexplored. We show how contextualized word embeddings can be used to capture affect dimensions in portrayals of people. We evaluate our methodology quantitatively, on held-out affect lexicons, and qualitatively, through case examples. We find that contextualized word representations do encode meaningful affect information, but they are heavily biased towards their training data, which limits their usefulness to in-domain analyses. We ultimately use our method to examine differences in portrayals of men and women.

Anjalie Field, Doron Kliger, Shuly Wintner et al.

Amidst growing concern over media manipulation, NLP attention has focused on overt strategies like censorship and "fake news'". Here, we draw on two concepts from the political science literature to explore subtler strategies for government media manipulation: agenda-setting (selecting what topics to cover) and framing (deciding how topics are covered). We analyze 13 years (100K articles) of the Russian newspaper Izvestia and identify a strategy of distraction: articles mention the U.S. more frequently in the month directly following an economic downturn in Russia. We introduce embedding-based methods for cross-lingually projecting English frames to Russian, and discover that these articles emphasize U.S. moral failings and threats to the U.S. Our work offers new ways to identify subtle media manipulation strategies at the intersection of agenda-setting and framing.

Anjalie Field

In the past several decades, many authorship attribution studies have used computational methods to determine the authors of disputed texts. Disputed authorship is a common problem in Classics, since little information about ancient documents has survived the centuries. Many scholars have questioned the authenticity of the final chapter of Xenophon's Cyropaedia, a 4th century B.C. historical text. In this study, we use N-grams frequency vectors with a cosine similarity function and word frequency vectors with Naive Bayes Classifiers (NBC) and Support Vector Machines (SVM) to analyze the authorship of the Cyropaedia. Although the N-gram analysis shows that the epilogue of the Cyropaedia differs slightly from the rest of the work, comparing the analysis of Xenophon with analyses of Aristotle and Plato suggests that this difference is not significant. Both NBC and SVM analyses of word frequencies show that the final chapter of the Cyropaedia is closely related to the other chapters of the Cyropaedia. Therefore, this analysis suggests that the disputed chapter was written by Xenophon. This information can help scholars better understand the Cyropaedia and also demonstrates the usefulness of applying modern authorship analysis techniques to classical literature.