I am at #NAACL2024 this week. Can we use LLM agents to simulate human-like society? TL;DR: They are not human-like by default, unless you explicitly induce human-like cognitive bias! Come chat with me! ⌚Wed (6/19) 11AM - 12:30PM

🚀 Excited to present our paper "Benchmarking Machine Translation with Cultural Awareness" at #EMNLP2024! We build CAMT, a novel parallel corpus enriched with culture-specific item annotations, and evaluate how well NMT and LLM-MT systems handle cultural entities.

Huge congratulations to Tim Rogers, recipient of the 2025 Jeffrey L. Elman Prize for Scientific Achievement and Community Building! This honor will be celebrated at #CogSci2025 with a prize and dedicated symposium. Visit the #ElmanPrize page for more cognitivesciencesociety.org/elman-prize/

Honored to have our paper "No Preference Left Behind: Group Distributional Preference Optimization (GDPO)" accepted to #ICLR2025! 🎉 Huge thanks to Binwei Yao for leading this work. TL;DR: Preference-based alignment methods like DPO eliminate minority opinions—how can we ensure

Excited to present our new LLM agent benchmark at the SEA @ #NeurIPS2025! 🎉 Paper: “DEBATE: A Large-Scale Benchmark for Role-Playing LLM Agents in Multi-Agent, Long-Form Debates” TL;DR: We release DEBATE, a large-scale empirical benchmark of 29k+ messages from 2,700+ U.S.