sail-sg /oat
🌾 OAT:一个面向研究的在线对齐框架,包括强化学习、偏好学习等。
Python675
llm-aligment 话题在 GitHub 上没有填写简介,这里按 Star 数汇总最相关的热门仓库。
🌾 OAT:一个面向研究的在线对齐框架,包括强化学习、偏好学习等。
A curated list of resources for activation engineering
official implementation of ICLR'2025 paper: Rethinking Bradley-Terry Models in Preference-based Reward Modeling: Foundations, Theory, and Alternatives
[NeurIPS 2024] Fast Best-of-N Decoding via Speculative Rejection
Some Thoughts on AI Alignment: Using AI to Control AI
Web application for benchmarking and comparing LLM behaviour, energy and emissions on cloud and bare metal under different inference configs, middleware scaffolds and post-training implementations