News Archives
- [12/2024] Our work featured our work on emergence of skill compositionality (and its limitations) in LLMs among the CS breakthroughs of the year Quanta Magazine!
- [09/2024] Congrats to Sadhika Malladi on being named 2025 Siebel Scholar!
- [09/2024] New paper Can Models Learn Skill Composition from Examples? was accepted by NeurIPS 2024!
- [09/2024] New paper Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates was accepted by NeurIPS 2024!
- [09/2024] New paper AI-Assisted Generation of Difficult Math Questions was accepted by MATH-AI Workshop at NeurIPS 2024!
- [05/2024] New paper Trainable transformer in transformer was accepted by ICML 2024!
- [05/2024] New paper LESS: Selecting Influential Data for Targeted Instruction Tuning was accepted by ICML 2024!
- [01/2024] Our work features our work on emergence of complex skills in LLMs and SkillMix evaluation Quanta Magazine!
- [01/2024] New paper Skill-Mix: A flexible and expandable family of evaluations for AI models was accepted by ICLR 2024!
- [09/2023] New paper Fine-Tuning Language Models with Just Forward Passes was accepted by NeurIPS 2023 Oral!
- [04/2023] New paper Task-Specific Skill Localization in Fine-tuned Language Models was accepted by ICML 2023!
- [04/2023] New paper A Kernel-Based View of Language Model Fine-Tuning was accepted by ICML 2023!
- [12/2020] Our work InstaHide: Instance-hiding Schemes for Private Distributed Learning is runner-up for Bell Labs Innovation prize!
- [09/2019] Andrej Risteski & Yuanzhi Li --> CMU, Nadav Cohen --> Tel Aviv.. Congratulations!
- [09/2017] Tengyu Ma --> Stanford, Yingyu Liang --> Madison. Congratulations!
- [09/2016] Tengyu Ma received best student paper at NIPS'16!