BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Diyi Yang

Diyi Yang contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Computation and LanguageArtificial IntelligenceMachine Learningcs.CYHuman-Computer InteractionSocial and Information NetworksComputer VisionCryptography and SecurityData Structures and AlgorithmsInformation RetrievalSoftware Engineering

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
34works
0followers
11topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

34 published item(s)

preprint2026arXiv

ProgramBench: Can Language Models Rebuild Programs From Scratch?

Turning ideas into full software projects from scratch has become a popular use case for language models. Agents are being deployed to seed, maintain, and grow codebases over extended periods with minimal human oversight. Such settings require models to make high-level software architecture decisions. However, existing benchmarks measure focused, limited tasks such as fixing a single bug or developing a single, specified feature. We therefore introduce ProgramBench to measure the ability of software engineering agents to develop software holisitically. In ProgramBench, given only a program and its documentation, agents must architect and implement a codebase that matches the reference executable's behavior. End-to-end behavioral tests are generated via agent-driven fuzzing, enabling evaluation without prescribing implementation structure. Our 200 tasks range from compact CLI tools to widely used software such as FFmpeg, SQLite, and the PHP interpreter. We evaluate 9 LMs and find that none fully resolve any task, with the best model passing 95\% of tests on only 3\% of tasks. Models favor monolithic, single-file implementations that diverge sharply from human-written code.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Reflections and New Directions for Human-Centered Large Language Models

Large Language Models (LLMs) are increasingly shaping the private and professional lives of users, with numerous applications in business, education, finance, healthcare, law, and science. With this rise in global influence comes greater urgency to build, evaluate, and deploy these systems in a manner that prioritizes not only technical capabilities but also human priorities. This work presents a framework for developing Human-Centered Large Language Models (HCLLMs), which integrates perspectives from Natural Language Processing (NLP), Human-Computer Interaction (HCI), and responsible AI. Considering the ethics, economics, and technical objectives of language modeling, we argue that model developers need to address human concerns, preferences, values, and goals, not only during a cursory post-training stage, but rather with rigor and care at every stage of the pipeline. This paper offers human-centered insights and recommendations for developers at each stage, from system design to data sourcing, model training, evaluation, and responsible deployment. Then we conclude with a case study, applying these insights to understand the future of work with HCLLMs.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Relative Scaling Laws for LLMs

Scaling laws describe how language models improve with additional data, parameters, and compute. While widely used, they are typically measured on aggregate test sets. Aggregate evaluations yield clean trends but average over heterogeneous subpopulations, obscuring performance disparities. We introduce relative scaling laws, which track how performance gaps between test distributions evolve with scale rather than focusing solely on absolute error. Using 255 decoder-only Transformers trained under matched-compute (IsoFLOP) budgets from $10^{18}$--$10^{20}$ FLOPs on standard pretraining datasets, we find diverse trajectories: academic domains on MMLU converge toward parity; regional English dialects shift depending on population size; and clusters of AI risk behaviours split, with capability- and influence-related risks increasing during pretraining while adversarial risks do not. These results show that although scaling improves overall performance, it is not a universal equalizer. To support further study, we release all model checkpoints from this work to enable practitioners to measure relative alongside traditional scaling laws, in order to better prioritize robustness challenges in light of the bitter lesson.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

SecureForge: Finding and Preventing Vulnerabilities in LLM-Generated Code via Prompt Optimization

LLM coding agents now generate code at an unprecedented scale, yet LLM-generated code introduces cybersecurity vulnerabilities into codebases without human involvement. Even when frontier models are explicitly asked to write secure production code with relevant weaknesses to avoid in context, we find that they still produce verifiable vulnerabilities on average 23% of the time across a corpus of 250 benign coding prompts. We introduce SecureForge, an automated pipeline that both audits security risks of frontier models and produces auditing-informed secure system prompts that reduce output security vulnerabilities while maintaining unit test performance. SecureForge first identifies benign prompts that produce statically detectable vulnerabilities, and then amplifies them into a large synthetic prompt corpus of diverse scenarios using a Markovian sampling technique to jointly maintain error rates and prompt diversity. This corpus is then used to iteratively optimize the system prompts to reduce output security vulnerabilities. On frontier models, SecureForge yields a statistically significant Pareto improvement in both unit test success and output security, with output vulnerabilities reduced by up to 48%. The resulting system prompts transfer zero-shot to in-the-wild coding agent prompts, without any exposure to real user prompt distributions during optimization.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Sycophantic AI makes human interaction feel more effortful and less satisfying over time

Millions of people now turn to artificial intelligence (AI) systems for personal advice, guidance, and support. Such systems can be sycophantic, frequently affirming users' views and beliefs. Across five preregistered studies (N = 3,075 participants, 12,766 human-AI conversations), including a three-week study with a census-representative U.S. sample, we provide longitudinal experimental evidence that sycophantic AI shifts how users approach their closest relationships. We show that sycophantic AI immediately delivers the emotional and esteem support users typically associate with close friends and family. Over three weeks of such interactions, users became nearly as likely to seek personal advice from sycophantic AI as from close friends and family, and reported lower satisfaction with their real-world social interactions. When given a choice among AI response styles, a majority preferred sycophantic AI -- not for the quality of its advice, but because it made them feel most understood. Together, these findings offer a relational account of AI sycophancy and its impacts.

0 citations0 reviewsNo code linkedOpen access
preprint2023arXiv

Identifying and Mitigating the Security Risks of Generative AI

Every major technical invention resurfaces the dual-use dilemma -- the new technology has the potential to be used for good as well as for harm. Generative AI (GenAI) techniques, such as large language models (LLMs) and diffusion models, have shown remarkable capabilities (e.g., in-context learning, code-completion, and text-to-image generation and editing). However, GenAI can be used just as well by attackers to generate new attacks and increase the velocity and efficacy of existing attacks. This paper reports the findings of a workshop held at Google (co-organized by Stanford University and the University of Wisconsin-Madison) on the dual-use dilemma posed by GenAI. This paper is not meant to be comprehensive, but is rather an attempt to synthesize some of the interesting findings from the workshop. We discuss short-term and long-term goals for the community on this topic. We hope this paper provides both a launching point for a discussion on this important topic as well as interesting problems that the research community can work to address.

0 citations0 reviewsNo code linkedOpen access
preprint2023arXiv

Parameter-Efficient Fine-Tuning Design Spaces

Parameter-efficient fine-tuning aims to achieve performance comparable to fine-tuning, using fewer trainable parameters. Several strategies (e.g., Adapters, prefix tuning, BitFit, and LoRA) have been proposed. However, their designs are hand-crafted separately, and it remains unclear whether certain design patterns exist for parameter-efficient fine-tuning. Thus, we present a parameter-efficient fine-tuning design paradigm and discover design patterns that are applicable to different experimental settings. Instead of focusing on designing another individual tuning strategy, we introduce parameter-efficient fine-tuning design spaces that parameterize tuning structures and tuning strategies. Specifically, any design space is characterized by four components: layer grouping, trainable parameter allocation, tunable groups, and strategy assignment. Starting from an initial design space, we progressively refine the space based on the model quality of each design choice and make greedy selection at each stage over these four components. We discover the following design patterns: (i) group layers in a spindle pattern; (ii) allocate the number of trainable parameters to layers uniformly; (iii) tune all the groups; (iv) assign proper tuning strategies to different groups. These design patterns result in new parameter-efficient fine-tuning methods. We show experimentally that these methods consistently and significantly outperform investigated parameter-efficient fine-tuning strategies across different backbone models and different tasks in natural language processing.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Search Engine for Discovery of Scientific Challenges and Directions

Keeping track of scientific challenges, advances and emerging directions is a fundamental part of research. However, researchers face a flood of papers that hinders discovery of important knowledge. In biomedicine, this directly impacts human lives. To address this problem, we present a novel task of extraction and search of scientific challenges and directions, to facilitate rapid knowledge discovery. We construct and release an expert-annotated corpus of texts sampled from full-length papers, labeled with novel semantic categories that generalize across many types of challenges and directions. We focus on a large corpus of interdisciplinary work relating to the COVID-19 pandemic, ranging from biomedicine to areas such as AI and economics. We apply a model trained on our data to identify challenges and directions across the corpus and build a dedicated search engine. In experiments with 19 researchers and clinicians using our system, we outperform a popular scientific search engine in assisting knowledge discovery. Finally, we show that models trained on our resource generalize to the wider biomedical domain and to AI papers, highlighting its broad utility. We make our data, model and search engine publicly available. https://challenges.apps.allenai.org/

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Sketch Is Worth a Thousand Words: Image Retrieval with Text and Sketch

We address the problem of retrieving images with both a sketch and a text query. We present TASK-former (Text And SKetch transformer), an end-to-end trainable model for image retrieval using a text description and a sketch as input. We argue that both input modalities complement each other in a manner that cannot be achieved easily by either one alone. TASK-former follows the late-fusion dual-encoder approach, similar to CLIP, which allows efficient and scalable retrieval since the retrieval set can be indexed independently of the queries. We empirically demonstrate that using an input sketch (even a poorly drawn one) in addition to text considerably increases retrieval recall compared to traditional text-based image retrieval. To evaluate our approach, we collect 5,000 hand-drawn sketches for images in the test set of the COCO dataset. The collected sketches are available a https://janesjanes.github.io/tsbir/.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyond

A fundamental goal of scientific research is to learn about causal relationships. However, despite its critical role in the life and social sciences, causality has not had the same importance in Natural Language Processing (NLP), which has traditionally placed more emphasis on predictive tasks. This distinction is beginning to fade, with an emerging area of interdisciplinary research at the convergence of causal inference and language processing. Still, research on causality in NLP remains scattered across domains without unified definitions, benchmark datasets and clear articulations of the challenges and opportunities in the application of causal inference to the textual domain, with its unique properties. In this survey, we consolidate research across academic areas and situate it in the broader NLP landscape. We introduce the statistical challenge of estimating causal effects with text, encompassing settings where text is used as an outcome, treatment, or to address confounding. In addition, we explore potential uses of causal inference to improve the robustness, fairness, and interpretability of NLP models. We thus provide a unified overview of causal inference for the NLP community.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Continual Sequence Generation with Adaptive Compositional Modules

Continual learning is essential for real-world deployment when there is a need to quickly adapt the model to new tasks without forgetting knowledge of old tasks. Existing work on continual sequence generation either always reuses existing parameters to learn new tasks, which is vulnerable to catastrophic forgetting on dissimilar tasks, or blindly adds new parameters for every new task, which could prevent knowledge sharing between similar tasks. To get the best of both worlds, in this work, we propose continual sequence generation with adaptive compositional modules to adaptively add modules in transformer architectures and compose both old and new modules for new tasks. We also incorporate pseudo experience replay to facilitate knowledge transfer in those shared modules. Experiment results on various sequences of generation tasks show that our framework can adaptively add modules or reuse modules based on task similarity, outperforming state-of-the-art baselines in terms of both performance and parameter efficiency. We make our code public at https://github.com/GT-SALT/Adaptive-Compositional-Modules.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

DoubleMix: Simple Interpolation-Based Data Augmentation for Text Classification

This paper proposes a simple yet effective interpolation-based data augmentation approach termed DoubleMix, to improve the robustness of models in text classification. DoubleMix first leverages a couple of simple augmentation operations to generate several perturbed samples for each training data, and then uses the perturbed data and original data to carry out a two-step interpolation in the hidden space of neural models. Concretely, it first mixes up the perturbed data to a synthetic sample and then mixes up the original data and the synthetic perturbed data. DoubleMix enhances models' robustness by learning the "shifted" features in hidden space. On six text classification benchmark datasets, our approach outperforms several popular text augmentation methods including token-level, sentence-level, and hidden-level data augmentation techniques. Also, experiments in low-resource settings show our approach consistently improves models' performance when the training data is scarce. Extensive ablation studies and case studies confirm that each component of our approach contributes to the final performance and show that our approach exhibits superior performance on challenging counterexamples. Additionally, visual analysis shows that text features generated by our approach are highly interpretable. Our code for this paper can be found at https://github.com/declare-lab/DoubleMix.git.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Fantastic Questions and Where to Find Them: FairytaleQA -- An Authentic Dataset for Narrative Comprehension

Question answering (QA) is a fundamental means to facilitate assessment and training of narrative comprehension skills for both machines and young children, yet there is scarcity of high-quality QA datasets carefully designed to serve this purpose. In particular, existing datasets rarely distinguish fine-grained reading skills, such as the understanding of varying narrative elements. Drawing on the reading education research, we introduce FairytaleQA, a dataset focusing on narrative comprehension of kindergarten to eighth-grade students. Generated by educational experts based on an evidence-based theoretical framework, FairytaleQA consists of 10,580 explicit and implicit questions derived from 278 children-friendly stories, covering seven types of narrative elements or relations. Our dataset is valuable in two folds: First, we ran existing QA models on our dataset and confirmed that this annotation helps assess models' fine-grained learning skills. Second, the dataset supports question generation (QG) task in the education domain. Through benchmarking with QG models, we show that the QG model trained on FairytaleQA is capable of asking high-quality and more diverse questions.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Graph Vulnerability and Robustness: A Survey

The study of network robustness is a critical tool in the characterization and sense making of complex interconnected systems such as infrastructure, communication and social networks. While significant research has been conducted in all of these areas, gaps in the surveying literature still exist. Answers to key questions are currently scattered across multiple scientific fields and numerous papers. In this survey, we distill key findings across numerous domains and provide researchers crucial access to important information by--(1) summarizing and comparing recent and classical graph robustness measures; (2) exploring which robustness measures are most applicable to different categories of networks (e.g., social, infrastructure; (3) reviewing common network attack strategies, and summarizing which attacks are most effective across different network topologies; and (4) extensive discussion on selecting defense techniques to mitigate attacks across a variety of networks. This survey guides researchers and practitioners in navigating the expansive field of network robustness, while summarizing answers to key questions. We conclude by highlighting current research directions and open problems.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Identifying and Mitigating Spurious Correlations for Improving Robustness in NLP Models

Recently, NLP models have achieved remarkable progress across a variety of tasks; however, they have also been criticized for being not robust. Many robustness problems can be attributed to models exploiting spurious correlations, or shortcuts between the training data and the task labels. Most existing work identifies a limited set of task-specific shortcuts via human priors or error analyses, which requires extensive expertise and efforts. In this paper, we aim to automatically identify such spurious correlations in NLP models at scale. We first leverage existing interpretability methods to extract tokens that significantly affect model's decision process from the input text. We then distinguish "genuine" tokens and "spurious" tokens by analyzing model predictions across multiple corpora and further verify them through knowledge-aware perturbations. We show that our proposed method can effectively and efficiently identify a scalable set of "shortcuts", and mitigating these leads to more robust models in multiple applications.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Inducing Positive Perspectives with Text Reframing

Sentiment transfer is one popular example of a text style transfer task, where the goal is to reverse the sentiment polarity of a text. With a sentiment reversal comes also a reversal in meaning. We introduce a different but related task called positive reframing in which we neutralize a negative point of view and generate a more positive perspective for the author without contradicting the original meaning. Our insistence on meaning preservation makes positive reframing a challenging and semantically rich task. To facilitate rapid progress, we introduce a large-scale benchmark, Positive Psychology Frames, with 8,349 sentence pairs and 12,755 structured annotations to explain positive reframing in terms of six theoretically-motivated reframing strategies. Then we evaluate a set of state-of-the-art text style transfer models, and conclude by discussing key challenges and directions for future work.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Know it to Defeat it: Exploring Health Rumor Characteristics and Debunking Efforts on Chinese Social Media during COVID-19 Crisis

Health-related rumors spreading online during a public crisis may pose a serious threat to people's well-being. Existing crisis informatics research lacks in-depth insights into the characteristics of health rumors and the efforts to debunk them on social media in a pandemic. To fill this gap, we conduct a comprehensive analysis of four months of rumor-related online discussion during COVID-19 on Weibo, a Chinese microblogging site. Results suggest that the dread (cause fear) type of health rumors provoked significantly more discussions and lasted longer than the wish (raise hope) type. We further explore how four kinds of social media users (i.e., government, media, organization, and individual) combat health rumors, and identify their preferred way of sharing debunking information and the key rhetoric strategies used in the process. We examine the relationship between debunking and rumor discussions using a Granger causality approach, and show the efficacy of debunking in suppressing rumor discussions, which is time-sensitive and varies across rumor types and debunkers. Our results can provide insights into crisis informatics and risk management on social media in pandemic settings.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Leveraging Expert Guided Adversarial Augmentation For Improving Generalization in Named Entity Recognition

Named Entity Recognition (NER) systems often demonstrate great performance on in-distribution data, but perform poorly on examples drawn from a shifted distribution. One way to evaluate the generalization ability of NER models is to use adversarial examples, on which the specific variations associated with named entities are rarely considered. To this end, we propose leveraging expert-guided heuristics to change the entity tokens and their surrounding contexts thereby altering their entity types as adversarial attacks. Using expert-guided heuristics, we augmented the CoNLL 2003 test set and manually annotated it to construct a high-quality challenging set. We found that state-of-the-art NER systems trained on CoNLL 2003 training data drop performance dramatically on our challenging set. By training on adversarial augmented training examples and using mixup for regularization, we were able to significantly improve the performance on the challenging set as well as improve out-of-domain generalization which we evaluated by using OntoNotes data. We have publicly released our dataset and code at https://github.com/GT-SALT/Guided-Adversarial-Augmentation.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Linguistic Characterization of Divisive Topics Online: Case Studies on Contentiousness in Abortion, Climate Change, and Gun Control

As public discourse continues to move and grow online, conversations about divisive topics on social media platforms have also increased. These divisive topics prompt both contentious and non-contentious conversations. Although what distinguishes these conversations, often framed as what makes these conversations contentious, is known in broad strokes, much less is known about the linguistic signature of these conversations. Prior work has shown that contentious content and structure can be a predictor for this task, however, most of them have been focused on conversation in general, very specific events, or complex structural analysis. Additionally, many models used in prior work have lacked interpret-ability, a key factor in online moderation. Our work fills these gaps by focusing on conversations from highly divisive topics (abortion, climate change, and gun control), operationalizing a set of novel linguistic and conversational characteristics and user factors, and incorporating them to build interpretable models. We demonstrate that such characteristics can largely improve the performance of prediction on this task, and also enable nuanced interpretability. Our case studies on these three contentious topics suggest that certain generic linguistic characteristics are highly correlated with contentiousness in conversations while others demonstrate significant contextual influences on specific divisive topics.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Measure and Improve Robustness in NLP Models: A Survey

As NLP models achieved state-of-the-art performances over benchmarks and gained wide applications, it has been increasingly important to ensure the safe deployment of these models in the real world, e.g., making sure the models are robust against unseen or challenging scenarios. Despite robustness being an increasingly studied topic, it has been separately explored in applications like vision and NLP, with various definitions, evaluation and mitigation strategies in multiple lines of research. In this paper, we aim to provide a unifying survey of how to define, measure and improve robustness in NLP. We first connect multiple definitions of robustness, then unify various lines of work on identifying robustness failures and evaluating models' robustness. Correspondingly, we present mitigation strategies that are data-driven, model-driven, and inductive-prior-based, with a more systematic view of how to effectively improve robustness in NLP models. Finally, we conclude by outlining open challenges and future directions to motivate further research in this area.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

SeqZero: Few-shot Compositional Semantic Parsing with Sequential Prompts and Zero-shot Models

Recent research showed promising results on combining pretrained language models (LMs) with canonical utterance for few-shot semantic parsing. The canonical utterance is often lengthy and complex due to the compositional structure of formal languages. Learning to generate such canonical utterance requires significant amount of data to reach high performance. Fine-tuning with only few-shot samples, the LMs can easily forget pretrained knowledge, overfit spurious biases, and suffer from compositionally out-of-distribution generalization errors. To tackle these issues, we propose a novel few-shot semantic parsing method -- SeqZero. SeqZero decomposes the problem into a sequence of sub-problems, which correspond to the sub-clauses of the formal language. Based on the decomposition, the LMs only need to generate short answers using prompts for predicting sub-clauses. Thus, SeqZero avoids generating a long canonical utterance at once. Moreover, SeqZero employs not only a few-shot model but also a zero-shot model to alleviate the overfitting. In particular, SeqZero brings out the merits from both models via ensemble equipped with our proposed constrained rescaling. SeqZero achieves SOTA performance of BART-based models on GeoQuery and EcommerceQuery, which are two few-shot datasets with compositional data split.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Six Feet Apart: Online Payments During the COVID-19 Pandemic

Since the COVID-19 pandemic, businesses have faced unprecedented challenges when trying to remain open. Because COVID-19 spreads through aerosolized droplets, businesses were forced to distance their services; in some cases, distancing may have involved moving business services online. In this work, we explore digitization strategies used by small businesses that remained open during the pandemic, and survey/interview small businesses owners to understand preliminary challenges associated with moving online. Furthermore, we analyze payments from 400K businesses across Japan, Australia, United States, Great Britain, and Canada. Following initial government interventions, we observe (at minimum for each country) a 47% increase in digitizing businesses compared to pre-pandemic levels, with about 80% of surveyed businesses digitizing in under a week. From both our quantitative models and our surveys/interviews, we find that businesses rapidly digitized at the start of the pandemic in preparation of future uncertainty. We also conduct a case-study of initial digitization in the United States, examining finer relationships between specific government interventions, business sectors, political orientation, and resulting digitization shifts. Finally, we discuss the implications of rapid & widespread digitization for small businesses in the context of usability challenges and interpersonal interactions, while highlighting potential shifts in pre-existing social norms.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

SUBS: Subtree Substitution for Compositional Semantic Parsing

Although sequence-to-sequence models often achieve good performance in semantic parsing for i.i.d. data, their performance is still inferior in compositional generalization. Several data augmentation methods have been proposed to alleviate this problem. However, prior work only leveraged superficial grammar or rules for data augmentation, which resulted in limited improvement. We propose to use subtree substitution for compositional data augmentation, where we consider subtrees with similar semantic functions as exchangeable. Our experiments showed that such augmented data led to significantly better performance on SCAN and GeoQuery, and reached new SOTA on compositional split of GeoQuery.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

The Moral Integrity Corpus: A Benchmark for Ethical Dialogue Systems

Conversational agents have come increasingly closer to human competence in open-domain dialogue settings; however, such models can reflect insensitive, hurtful, or entirely incoherent viewpoints that erode a user's trust in the moral integrity of the system. Moral deviations are difficult to mitigate because moral judgments are not universal, and there may be multiple competing judgments that apply to a situation simultaneously. In this work, we introduce a new resource, not to authoritatively resolve moral ambiguities, but instead to facilitate systematic understanding of the intuitions, values and moral judgments reflected in the utterances of dialogue systems. The Moral Integrity Corpus, MIC, is such a resource, which captures the moral assumptions of 38k prompt-reply pairs, using 99k distinct Rules of Thumb (RoTs). Each RoT reflects a particular moral conviction that can explain why a chatbot's reply may appear acceptable or problematic. We further organize RoTs with a set of 9 moral and social attributes and benchmark performance for attribute classification. Most importantly, we show that current neural language models can automatically generate new RoTs that reasonably describe previously unseen interactions, but they still struggle with certain scenarios. Our findings suggest that MIC will be a useful resource for understanding and language models' implicit moral assumptions and flexibly benchmarking the integrity of conversational agents. To download the data, see https://github.com/GT-SALT/mic

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

TreeMix: Compositional Constituency-based Data Augmentation for Natural Language Understanding

Data augmentation is an effective approach to tackle over-fitting. Many previous works have proposed different data augmentations strategies for NLP, such as noise injection, word replacement, back-translation etc. Though effective, they missed one important characteristic of language--compositionality, meaning of a complex expression is built from its sub-parts. Motivated by this, we propose a compositional data augmentation approach for natural language understanding called TreeMix. Specifically, TreeMix leverages constituency parsing tree to decompose sentences into constituent sub-structures and the Mixup data augmentation technique to recombine them to generate new sentences. Compared with previous approaches, TreeMix introduces greater diversity to the samples generated and encourages models to learn compositionality of NLP data. Extensive experiments on text classification and SCAN demonstrate that TreeMix outperforms current state-of-the-art data augmentation methods.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

VALUE: Understanding Dialect Disparity in NLU

English Natural Language Understanding (NLU) systems have achieved great performances and even outperformed humans on benchmarks like GLUE and SuperGLUE. However, these benchmarks contain only textbook Standard American English (SAE). Other dialects have been largely overlooked in the NLP community. This leads to biased and inequitable NLU systems that serve only a sub-population of speakers. To understand disparities in current models and to facilitate more dialect-competent NLU systems, we introduce the VernAcular Language Understanding Evaluation (VALUE) benchmark, a challenging variant of GLUE that we created with a set of lexical and morphosyntactic transformation rules. In this initial release (V.1), we construct rules for 11 features of African American Vernacular English (AAVE), and we recruit fluent AAVE speakers to validate each feature transformation via linguistic acceptability judgments in a participatory design manner. Experiments show that these new dialectal features can lead to a drop in model performance. To run the transformation code and download both synthetic and gold-standard dialectal GLUE benchmarks, see https://github.com/SALT-NLP/value

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Putting Humans in the Natural Language Processing Loop: A Survey

How can we design Natural Language Processing (NLP) systems that learn from human feedback? There is a growing research body of Human-in-the-loop (HITL) NLP frameworks that continuously integrate human feedback to improve the model itself. HITL NLP research is nascent but multifarious -- solving various NLP problems, collecting diverse feedback from different people, and applying different methods to learn from collected feedback. We present a survey of HITL NLP work from both Machine Learning (ML) and Human-Computer Interaction (HCI) communities that highlights its short yet inspiring history, and thoroughly summarize recent frameworks focusing on their tasks, goals, human interactions, and feedback learning methods. Finally, we discuss future directions for integrating human feedback in the NLP development loop.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

RECAST: Enabling User Recourse and Interpretability of Toxicity Detection Models with Interactive Visualization

With the widespread use of toxic language online, platforms are increasingly using automated systems that leverage advances in natural language processing to automatically flag and remove toxic comments. However, most automated systems -- when detecting and moderating toxic language -- do not provide feedback to their users, let alone provide an avenue of recourse for these users to make actionable changes. We present our work, RECAST, an interactive, open-sourced web tool for visualizing these models' toxic predictions, while providing alternative suggestions for flagged toxic language. Our work also provides users with a new path of recourse when using these automated moderation tools. RECAST highlights text responsible for classifying toxicity, and allows users to interactively substitute potentially toxic phrases with neutral alternatives. We examined the effect of RECAST via two large-scale user evaluations, and found that RECAST was highly effective at helping users reduce toxicity as detected through the model. Users also gained a stronger understanding of the underlying toxicity criterion used by black-box models, enabling transparency and recourse. In addition, we found that when users focus on optimizing language for these models instead of their own judgement (which is the implied incentive and goal of deploying automated models), these models cease to be effective classifiers of toxicity compared to human annotations. This opens a discussion for how toxicity detection models work and should work, and their effect on the future of online discourse.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Tuiteamos o pongamos un tuit? Investigating the Social Constraints of Loanword Integration in Spanish Social Media

Speakers of non-English languages often adopt loanwords from English to express new or unusual concepts. While these loanwords may be borrowed unchanged, speakers may also integrate the words to fit the constraints of their native language, e.g. creating Spanish "tuitear" from English "tweet." Linguists have often considered the process of loanword integration to be more dependent on language-internal constraints, but sociolinguistic constraints such as speaker background remain only qualitatively understood. We investigate the role of social context and speaker background in Spanish speakers' use of integrated loanwords on social media. We find first that newspaper authors use the integrated forms of loanwords and native words more often than social media authors, showing that integration is associated with formal domains. In social media, we find that speaker background and expectations of formality explain loanword and native word integration, such that authors who use more Spanish and who write to a wider audience tend to use integrated verb forms more often. This study shows that loanword integration reflects not only language-internal constraints but also social expectations that vary by conversation and speaker.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Weakly-Supervised Hierarchical Models for Predicting Persuasive Strategies in Good-faith Textual Requests

Modeling persuasive language has the potential to better facilitate our decision-making processes. Despite its importance, computational modeling of persuasion is still in its infancy, largely due to the lack of benchmark datasets that can provide quantitative labels of persuasive strategies to expedite this line of research. To this end, we introduce a large-scale multi-domain text corpus for modeling persuasive strategies in good-faith text requests. Moreover, we design a hierarchical weakly-supervised latent variable model that can leverage partially labeled data to predict such associated persuasive strategies for each sentence, where the supervision comes from both the overall document-level labels and very limited sentence-level labels. Experimental results showed that our proposed method outperformed existing semi-supervised baselines significantly. We have publicly released our code at https://github.com/GT-SALT/Persuasion_Strategy_WVAE.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Characterizing Collective Attention via Descriptor Context: A Case Study of Public Discussions of Crisis Events

Social media datasets make it possible to rapidly quantify collective attention to emerging topics and breaking news, such as crisis events. Collective attention is typically measured by aggregate counts, such as the number of posts that mention a name or hashtag. But according to rationalist models of natural language communication, the collective salience of each entity will be expressed not only in how often it is mentioned, but in the form that those mentions take. This is because natural language communication is premised on (and customized to) the expectations that speakers and writers have about how their messages will be interpreted by the intended audience. We test this idea by conducting a large-scale analysis of public online discussions of breaking news events on Facebook and Twitter, focusing on five recent crisis events. We examine how people refer to locations, focusing specifically on contextual descriptors, such as "San Juan" versus "San Juan, Puerto Rico." Rationalist accounts of natural language communication predict that such descriptors will be unnecessary (and therefore omitted) when the named entity is expected to have high prior salience to the reader. We find that the use of contextual descriptors is indeed associated with proxies for social and informational expectations, including macro-level factors like the location's global salience and micro-level factors like audience engagement. We also find a consistent decrease in descriptor context use over the lifespan of each crisis event. These findings provide evidence about how social media users communicate with their audiences, and point towards more fine-grained models of collective attention that may help researchers and crisis response organizations to better understand public perception of unfolding crisis events.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classification

This paper presents MixText, a semi-supervised learning method for text classification, which uses our newly designed data augmentation method called TMix. TMix creates a large amount of augmented training samples by interpolating text in hidden space. Moreover, we leverage recent advances in data augmentation to guess low-entropy labels for unlabeled data, hence making them as easy to use as labeled data.By mixing labeled, unlabeled and augmented data, MixText significantly outperformed current pre-trained and fined-tuned models and other state-of-the-art semi-supervised learning methods on several text classification benchmarks. The improvement is especially prominent when supervision is extremely limited. We have publicly released our code at https://github.com/GT-SALT/MixText.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

RECAST: Interactive Auditing of Automatic Toxicity Detection Models

As toxic language becomes nearly pervasive online, there has been increasing interest in leveraging the advancements in natural language processing (NLP), from very large transformer models to automatically detecting and removing toxic comments. Despite the fairness concerns, lack of adversarial robustness, and limited prediction explainability for deep learning systems, there is currently little work for auditing these systems and understanding how they work for both developers and users. We present our ongoing work, RECAST, an interactive tool for examining toxicity detection models by visualizing explanations for predictions and providing alternative wordings for detected toxic speech.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Semi-Supervised Models via Data Augmentationfor Classifying Interactive Affective Responses

We present semi-supervised models with data augmentation (SMDA), a semi-supervised text classification system to classify interactive affective responses. SMDA utilizes recent transformer-based models to encode each sentence and employs back translation techniques to paraphrase given sentences as augmented data. For labeled sentences, we performed data augmentations to uniform the label distributions and computed supervised loss during training process. For unlabeled sentences, we explored self-training by regarding low-entropy predictions over unlabeled sentences as pseudo labels, assuming high-confidence predictions as labeled data for training. We further introduced consistency regularization as unsupervised loss after data augmentations on unlabeled data, based on the assumption that the model should predict similar class distributions with original unlabeled sentences as input and augmented sentences as input. Via a set of experiments, we demonstrated that our system outperformed baseline models in terms of F1-score and accuracy.

0 citations0 reviewsNo code linkedOpen access