Christopher Potts
Professor of Linguistics
Studies computational semantics and pragmatics, and interpretability of language models.
Christopher Potts's lab at Stanford works on multilingual, translation & linguistics, large language models and trustworthy & safe ML. 71 papers since 2023, 25 at top venues, mostly EMNLP and ICLR. 11 PhD students. 48 alumni. Applications close .
At a glance
- papers, 2023–2026
- 19/ year
- at T1 venues
- 35%
- PhD students
- 11
- MS students
- 0
- alumni listed
- 48
Papers by yearfrom the lab homepage, 2023–2026
- 8 papers2026 · T1 4 · collecting
- 19 papers2025 · T1 9
- 25 papers2024 · T1 8
- 19 papers2023 · T1 4
T1 venuesOtherVenues: EMNLP, ICLR, ICML, Findings of EMNLP, NeurIPS
Research focuspaper titles that use each topic's words
- Large language models30 papers
- interpretability6 papers
- Multilingual, translation & linguistics3 papers
- semantics1 paper
- Trustworthy & safe ML0 papers
Paper list
From the lab homepage.
20268 papers
- ICLR (T1)Addressing divergent representations from causal interventions on neural networks (opens in a new tab)
- To appear in Proceedings of CoLMCounterfactual simulation training for chain-of-thought faithfulness (opens in a new tab)
- To appear in Proceedings of CoLMDo language models consistently encode the current year? (opens in a new tab)
- ICML (T1)Fast Byte Latent Transformer (opens in a new tab)
- ICLR (T1)GEPA: Reflective prompt evolution can outperform reinforcement learning (opens in a new tab)
- Commentary on Futrell and Mahowald , “How linguistics learned to stop worrying and love the language models”. Behavioral and Brain Sciences 49: e211Language models as tools for investigating the distinction between possible and impossible natural languages (opens in a new tab)
- ICML (T1)Outcome rewards do not guarantee verifiable or causally important reasoning (opens in a new tab)
- To appear in Proceedings of CoLMSynthetic data for any differentiable target (opens in a new tab)
202519 papers
- Transactions of the ACL 13: 442-460Anchored Preference Optimization and Contrastive Revisions: Addressing underspecification in alignment (opens in a new tab)
- ICML (T1)AxBench: Steering LLMs? Even simple baselines outperform sparse autoencoders (opens in a new tab)
- In Proceedings of CoLMBase models beat aligned models at randomness and creativity (opens in a new tab)
- In Proceedings of CoLMBayesian scaling laws for in-context learning (opens in a new tab)
- In Advances in Neural Information Processing Systems 38: 162337-162367. Curran Associates, IncBlackbox model provenance via palimpsestic membership inference (opens in a new tab)
- Journal of Machine Learning Research 26(83): 1-64Causal abstraction: A theoretical foundation for mechanistic interpretability (opens in a new tab)
- EMNLP (T1)Causal interventions reveal shared structure across English filler-gap constructions (opens in a new tab)
- In Advances in Information Retrieval: European Conference on Information Retrieval, Part IV, 21-30. Springer-VerlagColBERT-Serve: Efficient multi-stage memory-mapped scoring (opens in a new tab)
- In Proceedings of the ACM Conference on AI and Agentic Systems, 197-213. Association for Computing MachineryComposing policy gradients and prompt optimization for language model programs (opens in a new tab)
- Findings of EMNLPDistinguishing fair from unfair compositional generalization tasks (opens in a new tab)
- NeurIPS (T1)Do language models use their depth efficiently? (opens in a new tab)
- Findings of EMNLPFalse friends are not foes: Investigating vocabulary overlap in multilingual language models (opens in a new tab)
202425 papers
- In Proceedings of the Conference on Empirical Methods in Natural Language Processing: Industry Track, 1046-1055. Miami, Florida: Association for Computational LinguisticsAmazonQAC: A large-scale, naturalistic query autocomplete dataset (opens in a new tab)
- In Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), 338-354. Mexico City: Association for Computational LinguisticsARES: An automated evaluation framework for retrieval-augmented generation systems (opens in a new tab)
- Knowledge-Based Systems 203: 112243Building efficient and effective OpenQA systems for low-resource languages (opens in a new tab)
- In Proceedings of the Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 14638-14663. Bangkok, Thailand: Association for Computational LinguisticsCausalGym: Benchmarking causal interpretability methods on linguistic tasks (opens in a new tab)
- Journal of Linguistics. Published online :1-39Characterizing English Preposing in PP constructions (opens in a new tab)
- EMNLP (T1)CommVQA: Situating visual question answering in communicative contexts (opens in a new tab)
- In Michael C. Frank and Asifa Majid, eds., Open Encyclopedia of Cognitive Science. MIT PressCompositionality (opens in a new tab)
- ICLR (T1)ContextRef: Evaluating referenceless metrics for image description generation (opens in a new tab)
- EMNLP (T1)Demystifying verbatim memorization in large language models (opens in a new tab)
- ICLR (T1)DSPy: Compiling declarative language model calls into self-improving pipelines (opens in a new tab)
- Proceedings of the Annual Meeting of the Cognitive Science SocietyEvaluating human and machine understanding of data visualizations (opens in a new tab)
- EMNLP (T1)Fine-tuning and prompt optimization: Two great steps that work better together (opens in a new tab)
202319 papers
- Medium postAgainst University-wide policies concerning AI writing assistants (opens in a new tab)
- Findings of EMNLPBioDEX: Large-scale biomedical adverse drug event extraction for real-world pharmacovigilance (opens in a new tab)
- ICML (T1)Causal Proxy Models for concept-based model explanations (opens in a new tab)
- In Proceeedings of the Sixth Workshop on Computational Models of Reference, Anaphora and CoreferenceCAW-coref: Conjunction-Aware Word-level coreference resolution (opens in a new tab)
- Wiki WorkshopCharacterizing image accessibility on Wikipedia across languages (opens in a new tab)
- ICCV WorkshopContext-VQA: Towards context-aware and purposeful Visual Question Answering (opens in a new tab)
- In Proceedings of the Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), 1803-1821. Toronto: Association for Computational LinguisticsDetecting contradictory COVID-19 drug efficacy claims from biomedical literature (opens in a new tab)
- In Proceedings of Causal Learning and ReasoningFinding alignments between interpretable causal variables and distributed neural representations (opens in a new tab)
- In BlackBoxNLP . Singapore: Association for Computational LinguisticsFlexible model interpretability through natural language model editing (opens in a new tab)
- Findings of ACLInducing character-level structure in subword-based language models with Type-level Interchange Intervention Training (opens in a new tab)
- NeurIPS (T1)Interpretability at scale: Identifying causal mechanisms in Alpaca (opens in a new tab)
- Findings of EACLLexical semantics with Large Language Models: A case study of English break (opens in a new tab)
Where alumni went48 listed
Apply to StanfordFall 2027
Stanford Computer Science PhD
- Round
- Single round
- Opens
- GRE
- See program page
- English
- TOEFL/IELTS
- Fee
- See program page
Confirmed on the program page.
Similar labsnearest by research summary
- CNSJTU
Yun Lin
- Large language models
- GBUCL
Yao Lu
- Large language models
- USStanford
Yejin Choi
- Large language models
- GBEdinburgh
Shay Cohen
- Multilingual, translation & linguistics
- GBCambridge
Nigel Collier
- Large language models
- GBEdinburgh
Alexandra Birch
- Multilingual, translation & linguistics
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
Stanfordcloses