
Yunyi Shen/็ณไบ้ธ ๐บ
@shenraphael
I do random statistics stuff at @MIT EECS. Once UW-Madison and PKU. Take wildlife photos. Opinions are mine.
ID: 1288542460831043585
http://yunyishen.github.io 29-07-2020 18:30:44
2,2K Tweet
2,2K Takipรงi
236 Takip Edilen




๐จPast work shows: dropping just 0.1% of the data can change the conclusions of important studies. We show: Many approximations can fail to catch this. ๐ขCheck out our new TMLR paper (w/ David Burt, Yunyi Shen/็ณไบ้ธ ๐บ , Tin Nguyen, and Tamara Broderick ) ๐ openreview.net/forum?id=m6EQ6โฆ


๐จDropping Just a Handful of Preferences Can Change Top Large Language Model Rankings. ๐ Oral @ ICML 2025 Workshop on Models of Human Feedback for AI Alignment ๐New work w/ Yunyi Shen/็ณไบ้ธ ๐บ, Dennis Wei, and Tamara Broderick openreview.net/forum?id=b9r1sโฆ





๐ RL is powering breakthroughs in LLM alignment, reasoning, and agentic apps. Are you ready to dive into the RL x LLM frontier? Join us at ACL 2025 ACLโ25 tutorial: Inverse RL Meets LLM Alignment this Sunday at Vienna๐ฆ๐น(Jul 27th, 9am) ๐ Preprint at huggingface.co/papers/2507.13โฆ







alex peysakhovich ๐ค Adam Lerer You missed a crucial piece between argue and ship: The main missing idea comes in the hangover shower!
