Loading activity…
@maartensap
retiring X acct: find me @maartensap.bsky Working on #NLProc for social good. Currently at @LTIatCMU, previously at @UWNLP, @MSFTResearch, and @allen_ai. 🏳🌈
AI
The paper’s arXiv description says general-purpose AI such as ChatGPT seems to have lowered barriers to public use.
AI
A user calls a quoted post “great ideas” for things to have agents do.
AI
The author said COLM 2026 was still underway and invited attendees to see the paper in the Imperial Ballroom.
AI
An October 6 thread lists Xuhui Zhou and Mingqian Zheng as the respective presenters.
AI
October 6 posts listed GoodPoint and Social World Models for Session 4 at 4:30 p.m. PDT.
AI
Posts list Yashwanth YS presenting SOTOPIA-TOM at 11 a.m. PDT and Jimin Mun presenting GoodPoint at 4:30 p.m. PDT.
AI
An October 6 thread lists Akhila Yerukola and Yashwanth YS as presenters for Session 3 at 11 AM PDT.
AI
A post introduces CCBench as an assessment of LLM cultural competence, led and presented by Vasudha Varadarajan.
AI
Posts list Xuhui Zhou as leading an 11 a.m. PDT talk and Zhou with Weiwei Sun as leading a 4:30 p.m. talk.
AI
Maarten Sap listed presentations by Jocelyn Shen and Xuhui Zhou for Session 1 at 11 a.m. PDT.
AI
The team behind CC-Bench says its new benchmark for LLM cultural competence was accepted to COLM 2026.
AI
Ai2 tool audits whether benchmark questions test intended model abilities such as safety.
