Advancing Science- and Evidence-based AI Policy
Comments This is the author's version of the work. It is posted here by permission of the AAAS for personal use, not for redistribution. The definitive version was published in Science on July 31, 2025
作者
Dan Jurafsky
Natural Language Processing
What can large language models do for sustainable food?
机构 * Stanford University（斯坦福大学）
Comments ICML camera ready version
HumT DumT: Measuring and controlling human-like language in LLMs
机构 * Stanford University（斯坦福大学）
Comments Accepted to ACL 2025
In-Context Learning Boosts Speech Recognition via Human-like Adaptation to Speakers and Language Varieties
机构 * Stanford University（斯坦福大学） ； University of Cambridge（剑桥大学）
Comments 15 pages; 3 figures
Labeling Messages as AI-Generated Does Not Reduce Their Persuasive Effects
机构 * Department of Computer Science, Stanford University（计算机科学系，斯坦福大学） ； Stanford Law School, Stanford University（斯坦福法学院，斯坦福大学） ； Department of Computer Science, Northeastern University（计算机科学系，东北大学） ； Department of Sociology, Stanford University（社会学系，斯坦福大学）
Dynamic Cheatsheet: Test-Time Learning with Adaptive Memory
h4rm3l: A language for Composable Jailbreak Attack Synthesis
Comments Accepted to the Thirteenth International Conference on Learning Representations (ICLR 2025)
AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders
Rethinking Word Similarity: Semantic Similarity through Classification Confusion
Comments Accepted to NAACL-main-2025
Can Unconfident LLM Annotations Be Used for Confident Conclusions?
Comments Please cite as: Can Unconfident LLM Annotations Be Used for Confident Conclusions? Kristina Gligorić, Tijana Zrnic, Cinoo Lee, Emmanuel Candès, and Dan Jurafsky. NAACL, 2025
Using large language models to promote health equity
KTO: Model Alignment as Prospect Theoretic Optimization
Comments ICML 2024
Belief in the Machine: Investigating Epistemological Blind Spots of Language Models
Comments https://github.com/suzgunmirac/belief-in-the-machine
Rel-A.I.: An Interaction-Centered Approach To Measuring Human-LM Reliance
Comments Preprint
A layer-wise analysis of Mandarin and English suprasegmentals in SSL speech models
Comments 4 pages, 3 figures, to be published in Interspeech 2024 proceedings
Data Checklist: On Unit-Testing Datasets with Usable Information
Comments 17 pages, 4 figures. COLM 2024
ML-SUPERB 2.0: Benchmarking Multilingual Speech Models Across Modeling Constraints, Languages, and Datasets
Comments Accepted by Interspeech 2024
ReFT: Representation Finetuning for Language Models
Comments preprint
Learning the meanings of function words from grounded language using a visual question answering model
Comments Published in Cognitive Science 2024
Ecosystem-level Analysis of Deployed Machine Learning Reveals Homogeneous Outcomes
Comments Published at NeurIPS 2023. Code is available at https://github.com/rishibommasani/EcosystemLevelAnalysis
Grounding Gaps in Language Model Generations
Comments NAACL 2024; 18 pages, 2 figures
NLP Systems That Can't Tell Use from Mention Censor Counterspeech, but Teaching the Distinction Helps
Comments NAACL 2024 (Main conference)
Othering and low status framing of immigrant cuisines in US restaurant reviews and large language models
Comments ICWSM '24
Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions
Dialect prejudice predicts AI decisions about people's character, employability, and criminality
CausalGym: Benchmarking causal interpretability methods on linguistic tasks
Comments 9 pages main text, 26 pages total
How Well Can LLMs Negotiate? NegotiationArena Platform and Analysis
A Benchmark for Learning to Translate a New Language from One Grammar Book
Comments Project site: https://lukemelas.github.io/mtob/
Predicting positive transfer for improved low-resource speech recognition using acoustic pseudo-tokens
Comments Accepted for SIGTYP2024
AnthroScore: A Computational Linguistic Measure of Anthropomorphism
Comments EACL 2024 Main Conference
