Lucy Li

@lucy3_li

Assistant prof @WisconsinCS. Prev @UCBerkeley, @allen_ai, @MSFTResearch, @stanfordnlp, @uwnlp. More silly at http://lucy3.bsky.social.

following
followers

humans& announces Ampersand Tutor for its agent API

humans& reports a 0.754 TutorMoments score under a plain prompt, versus 0.500 for its GLM-5.3 starting model.

44 Sources, 10d ago, updated 10d ago

Audio language model paper reportedly wins EMNLP 2026 outstanding paper award

In an earlier post, @lucy3_li described the research as examining unequal treatment of English varieties in high-stakes settings.

8 Sources, 12d ago, updated 12d ago

People may still believe false content despite language-model pushback

A paper co-author says models' challenges to faulty inputs are ambiguous and ineffective.

1 Source, 27d ago, first seen 27d ago

A screenshot of the top half of the first page of a paper titled "Pushback in a Haystack: Language Models Respond to Factual Errors Ambiguously and Ineffectively" with authors Li Lucy, Myra Cheng, Svenja Guhr, Kabir Ahuja, Orevaoghene Ahia, Sameep Chattopadhyay, Yike Wang, Valentin Hofmann, Noah A. Smith, Yulia Tsvetkov

X users congratulated Nathan Lambert and Trillium Labs after a launch, praising an open-science approach

Posts on X congratulated Nathan Lambert and Trillium Labs after a launch, with several users describing the effort as open, science-oriented or focused on open post-training. Lambert later said some criticism came from Substack and more general-audience AI followers, not the technical AI community.