Christopher Potts

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders ICML 2025 Causal Abstraction: A Theoretical Foundation for Mechanistic Interpretability JMLR 2025 Causal Interventions Reveal Shared Structure Across English Filler–Gap Constructions EMNLP 2025 Distinguishing fair from unfair compositional generalization tasks EMNLP 2025 False Friends Are Not Foes: Investigating Vocabulary Overlap in Multilingual Language Models EMNLP 2025 HyperDAS: Towards Automating Mechanistic Interpretability with Hypernetworks ICLR 2025 Improving Pretraining Data Using Perplexity Correlations ICLR 2025 Internal Causal Mechanisms Robustly Predict Language Model Out-of-Distribution Behaviors ICML 2025 MrT5: Dynamic Token Merging for Efficient Byte-level Language Models ICLR 2025 AmazonQAC: A Large-Scale, Naturalistic Query Autocomplete Dataset EMNLP 2024 ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems NAACL 2024 CausalGym: Benchmarking causal interpretability methods on linguistic tasks ACL 2024 CommVQA: Situating Visual Question Answering in Communicative Contexts EMNLP 2024 ContextRef: Evaluating Referenceless Metrics for Image Description Generation ICLR 2024 Demystifying Verbatim Memorization in Large Language Models EMNLP 2024 DSPy: Compiling Declarative Language Model Calls into State-of-the-Art Pipelines ICLR 2024 Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations CLeaR 2024 Fine-Tuning and Prompt Optimization: Two Great Steps that Work Better Together EMNLP 2024 GIO: Gradient Information Optimization for Training Dataset Selection ICLR 2024 I am a Strange Dataset: Metalinguistic Tests for Language Models ACL 2024 Mission: Impossible Language Models ACL 2024 MoEUT: Mixture-of-Experts Universal Transformers NeurIPS 2024 MSCAW-coref: Multilingual, Singleton and Conjunction-Aware Word-Level Coreference Resolution EMNLP 2024 Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programs EMNLP 2024 pyvene: A Library for Understanding and Improving PyTorch Models via Interventions NAACL 2024 RAVEL: Evaluating Interpretability Methods on Disentangling Language Model Representations ACL 2024 Recurrent Neural Networks Learn to Store and Generate Sequences using Non-Linear Representations EMNLP 2024 ReFT: Representation Finetuning for Language Models NeurIPS 2024 Retrieval Augmented Spelling Correction for E-Commerce Applications EMNLP 2024 Updating CLIP to Prefer Descriptions Over Captions EMNLP 2024 BioDEX: Large-Scale Biomedical Adverse Drug Event Extraction for Real-World Pharmacovigilance EMNLP 2023 Causal Proxy Models for Concept-based Model Explanations ICML 2023 CAW-coref: Conjunction-Aware Word-level Coreference Resolution EMNLP 2023 Detecting Contradictory COVID-19 Drug Efficacy Claims from Biomedical Literature ACL 2023 Inducing Character-level Structure in Subword-based Language Models with Type-level Interchange Intervention Training ACL 2023 Interpretability at Scale: Identifying Causal Mechanisms in Alpaca NeurIPS 2023 Lexical Semantics with Large Language Models: A Case Study of English “break” EACL 2023 Moving Beyond Downstream Task Accuracy for Information Retrieval Benchmarking ACL 2023 MQuAKE: Assessing Knowledge Editing in Language Models via Multi-Hop Questions EMNLP 2023 Multi-teacher Distillation for Multilingual Spelling Correction EMNLP 2023 Rigorously Assessing Natural Language Explanations of Neurons EMNLP 2023 ScoNe: Benchmarking Negation Reasoning in Language Models With Fine-Tuning and In-Context Learning ACL 2023 UDAPDR: Unsupervised Domain Adaptation via LLM Prompting and Distillation of Rerankers EMNLP 2023 Causal Distillation for Language Models NAACL 2022 CEBaB: Estimating the Causal Effects of Real-World Concepts on NLP Model Behavior NeurIPS 2022 ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction NAACL 2022 Concadia: Towards Image-Based Text Generation with a Purpose EMNLP 2022 Context Matters for Image Descriptions for Accessibility: Challenges for Referenceless Evaluation Metrics EMNLP 2022 Hindsight: Posterior-guided training of retrievers for improved open-ended generation ICLR 2022 Identifying the Limits of Cross-Domain Knowledge Transfer for Pretrained Models ACL 2022 Inducing Causal Structure for Interpretable Neural Networks ICML 2022 Systematicity in GPT-3’s Interpretation of Novel English Noun Compounds EMNLP 2022 Baleen: Robust Multi-Hop Reasoning at Scale via Condensed Retrieval NeurIPS 2021 Causal Abstractions of Neural Networks NeurIPS 2021 Decrypting Cryptic Crosswords: Semantically Complex Wordplay Puzzles as a Target for NLP NeurIPS 2021 Dynabench: Rethinking Benchmarking in NLP NAACL 2021 Dynaboard: An Evaluation-As-A-Service Platform for Holistic Next-Generation Benchmarking NeurIPS 2021 DynaSent: A Dynamic Benchmark for Sentiment Analysis ACL 2021 Data and Representation for Turkish Natural Language Inference EMNLP 2020 Modeling Subjective Assessments of Guilt in Newspaper Crime Narratives CoNLL 2020 Neural Natural Language Inference Models Partially Embed Theories of Lexical Entailment and Negation EMNLP 2020 Pragmatic Issue-Sensitive Image Captioning EMNLP 2020 Effective Feature Representation for Clinical Text Concept Extraction NAACL 2019 Posing Fair Generalization Tasks for Natural Language Inference EMNLP 2019 Recursive Routing Networks: Learning to Compose Modules for Language Understanding NAACL 2019 TalkDown: A Corpus for Condescension Detection in Context EMNLP 2019 Generating Bilingual Pragmatic Color References NAACL 2018 Mittens: an Extension of GloVe for Learning Domain-Specialized Representations NAACL 2018 Pragmatically Informative Image Captioning with Character-Level Inference NAACL 2018 Representing Social Media Users for Sarcasm Detection EMNLP 2018 Retrofitting Distributional Embeddings to Knowledge Graphs with Functional Relations COLING 2018 A Fast Unified Model for Parsing and Sentence Understanding ACL 2016 Learning to Generate Compositional Color Descriptions EMNLP 2016 A large annotated corpus for learning natural language inference EMNLP 2015 Text to 3D Scene Generation with Rich Lexical Grounding ACL 2015 A computational approach to politeness with application to social factors ACL 2013 Emergence of Gricean Maxims from Multi-Agent Decision Theory NAACL 2013 Implicatures and Nested Beliefs in Approximate Decentralized-POMDPs ACL 2013 Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank EMNLP 2013 The Life and Death of Discourse Entities: Identifying Singleton Mentions NAACL 2013 Learning Word Vectors for Sentiment Analysis ACL 2011 “Was It Good? It Was Provocative.” Learning the Meaning of Scalar Adjectives ACL 2010