News:
- We've developed an AI that plays 🧛♀️ Werewolf game as skillfully as humans do! Dive into our training methodology in our latest paper on 👾 Multi-agent KTO.
- 🥳 Our paper AgentSense is accepted by NAACL 2024 main conference!
- 🧐 How can we construct an automated debate judge to evaluate an extensive, vibrant, multi-turn debate? We designed Debatrix, a novel framework for automated debate judging based on LLM, featuring a vertical, iterative chronological analysis and a horizontal, multi-dimensional evaluation collaboration. Check out our paper ACL 2024.
- 🥳 Two papers on AI debate, Argue with me tersely and Hi-ArG, are accepted by EMNLP 2023!
- 🤖 Happy to share our industrial practice on speech translation with LLM. -- Speech Translation with Large Language Models: An Industrial Practice and demo
|
