Yulia Tsvetkov
Associate Professor
Studies multilingual and socially aware NLP, including fairness and safety of language models.
Yulia Tsvetkov's lab at UW works on interpretability, evaluation & safety of LMs and large language models. 89 papers since 2023, 67 at top venues, mostly ACL and EMNLP. 10 PhD students. 20 alumni, most often to Amazon and Google. Applications close .
At a glance
- papers, 2023–2026
- 23/ year
- at T1 venues
- 75%
- PhD students
- 10
- MS students
- 0
- alumni listed
- 20
Papers by yearfrom the lab homepage, 2023–2026
- 19 papers2026 · T1 14 · collecting
- 18 papers2025 · T1 14
- 34 papers2024 · T1 29
- 18 papers2023 · T1 10
T1 venuesOtherVenues: ACL, EMNLP, NeurIPS, NAACL, ICLR
Research focuspaper titles that use each topic's words
- Large language models56 papers
- Interpretability, evaluation & safety of LMs4 papers
- multilingual3 papers
- fairness1 paper
- socially0 papers
Paper list
From the lab homepage.
202619 papers
- Proc. SatMLA False Sense of Privacy: Evaluating Textual Data Sanitization Beyond Surface-level Privacy Leakage (opens in a new tab)
- Proc. COLMAds in AI Chatbots? An Analysis of How Large Language Models Navigate Conflicts of Interest (opens in a new tab)
- ACL (T1)Among Us: Measuring and Mitigating Malicious Contributions in Model Collaboration Systems (opens in a new tab)
- ICML (T1)Cold-Start Personalization via Training-Free Priors from Structured World Models (opens in a new tab)
- ACL (T1)Data Swarms: Optimizable Generation of Synthetic Evaluation Data (opens in a new tab)
- ACL (T1)Generalizable LLM Learning of Graph Synthetic Data with Post-training Alignment (opens in a new tab)
- IUIInteractive Reasoning: Visualizing and Controlling Chain-of-Thought Reasoning in Large Language Models (opens in a new tab)
- ICLR (T1)MADFormer: Mixed Autoregressive and Diffusion Transformers for Continuous Image Generation (opens in a new tab)
- Proc. COLMMMMG: A Comprehensive and Reliable Benchmark for Multitask Multimodal Generation (opens in a new tab)
- EMNLP (T1)Moral Safety in LLMs: Exposing Performative Compliance with Puzzled Cues (opens in a new tab)
- EMNLP (T1)One Form to Transfer Them All: Pretraining Multilingual Language Models Beyond Native Orthography (opens in a new tab)
- ICLR (T1)PrefDisco: Benchmarking Proactive Personalized Reasoning (opens in a new tab)
202518 papers
- Proc. COLMALFA: Aligning LLMs to Ask Good Questions A Case Study in Clinical Reasoning (opens in a new tab)
- NAACL (T1)Alpaca against Vicuna: Using LLMs to Uncover Memorization of LLMs (opens in a new tab)
- NeurIPS (T1)Artificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond) (opens in a new tab)
- ACL (T1)Biased AI can Influence Political Decision-Making (opens in a new tab)
- NAACL (T1)ComPO: Community Preferences for Language Model Personalization (opens in a new tab)
- ACL (T1)CulturalBench: A Robust, Diverse, and Challenging Cultural Benchmark by Human-AI CulturalTeaming (opens in a new tab)
- NeurIPS (T1)Escaping the SpuriVerse: Can Large Vision-Language Models Generalize Beyond Seen Spurious Correlations? (opens in a new tab)
- ICLR (T1)Explore Theory of Mind: Program-guided adversarial data generation for theory of mind reasoning (opens in a new tab)
- NAACL WorkshopFacts&Evidence: An Interactive Tool for Transparent Fine-Grained Factual Verification of Machine-Generated Text (opens in a new tab)
- Proc. COLMFinding Flawed Fictions: Evaluating Complex Reasoning in Language Models via Plot Hole Detection (opens in a new tab)
- NeurIPS (T1)Heterogeneous Swarms: Jointly Optimizing Model Roles and Weights for Multi-LLM Systems (opens in a new tab)
- TACL (T1)Know Your Limits: A Survey of Abstention in Large Language Models (opens in a new tab)
202434 papers
- NAACL (T1)BUFFET: Benchmarking Large Language Models for Few-shot Cross-lingual Transfer (opens in a new tab)
- EMNLP (T1)Can LLM Graph Reasoning Generalize beyond Pattern Memorization? (opens in a new tab)
- ICLR (T1)Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory (opens in a new tab)
- Nature Machine Intelligence (T1)Can Machines Learn Morality? The Delphi Experiment (opens in a new tab)
- NAACL (T1)David helps Goliath: Inference-Time Collaboration Between Small Specialized and Large General Diffusion LMs (opens in a new tab)
- ACL (T1)DELL: Generating Reactions and Explanations for LLM-Based Misinformation Detection (opens in a new tab)
- ACL (T1)DIALECTBENCH: A NLP Benchmark for Dialects, Varieties, and Closely-Related Languages (opens in a new tab)
- Proc. COLMDo Membership Inference Attacks Work on Large Language Models? (opens in a new tab)
- ACL (T1)Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaboration (opens in a new tab)
- NAACL (T1)Extracting Lexical Features from Dialects via Interpretable Dialect Classifiers (opens in a new tab)
- Proc. COLMFine-grained Hallucination Detection and Editing for Language Models (opens in a new tab)
- ICLR (T1)Gen-Z: Generative Zero-Shot Text Classification with Contextualized Label Descriptions (opens in a new tab)
202318 papers
- Findings of EMNLPBotPercent: Estimating Twitter Bot Populations from Groups to Crowds (opens in a new tab)
- NeurIPS (T1)Can Language Models Solve Graph Problems in Natural Language? (opens in a new tab)
- EMNLP (T1)Do All Languages Cost the Same? Tokenization in the Era of Commercial Language Models (opens in a new tab)
- FAccTExamining Risks of Racial Biases in NLP Tools for Child Protective Services (opens in a new tab)
- EMNLP (T1)FactKB: Generalizable Factuality Evaluation using Language Models Enhanced with Factual Knowledge (opens in a new tab)
- ACL (T1)From Pretraining Data to Language Models to Downstream Tasks: Tracking the Trails of Political Biases Leading to Unfair NLP Models (opens in a new tab)
- EMNLP (T1)GlobalBench: A Benchmark for Global Progress in Natural Language Processing (opens in a new tab)
- ACL (T1)KALM: Knowledge-Aware Integration of Local, Document, and Global Contexts for Long Document Understanding (opens in a new tab)
- EACLLanguage Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey (opens in a new tab)
- Proc. StarSEMLEXPLAIN: Improving Model Explanations via Lexicon Supervision (opens in a new tab)
- ACL (T1)Minding Language Models' (Lack of) Theory of Mind: A Plug-and-Play Multi-Character Belief Tracker (opens in a new tab)
- ACL (T1)On the Blind Spots of Model-Based Evaluation Metrics for Text Generation (opens in a new tab)
Where alumni went20 listed, 50% to industry
Industry 9Academia 9
Most often to
- Amazon
- Meta
Also to
- Stanford University
- Carnegie Mellon University
- Tsinghua University
Apply to UWFall 2027
UW Allen School Computer Science PhD
- Round
- Single round
- Opens
- GRE
- not required
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USStanford
Monica Lam
- Large language models
- GBUCL
He Ye
- Large language models
- USCMU
Alon Lavie
- Interpretability, evaluation & safety of LMs
- CNTsinghua
Zhijian Ou
- Speech recognition & spoken language
- CNZhejiang
Ming Cai
0 T1 - CNSJTU
Sufeng Duan
- Large language models
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.
UWcloses