← Trending

Tangerine Index lists products launched since ChatGPT and people's reactions to them.

reward-bench

github.com

RewardBench: the first evaluation tool for reward models.

Open source

Visit github.comexternal siteGitHub749 starsexternal site

Reactions (1)

Similar products

by description similarity

  1. RLHF-Reward-Modeling

    Recipes to train reward model for RLHF.

    GitHub1,546 starsgithub.comOpen sourcelatest reaction

  2. rlhf-book

    Textbook on reinforcement learning from human feedback

    GitHub2,514 starsgithub.comOpen sourcelatest reaction

  3. OpenJudge

    OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards

    GitHub1,080 starsgithub.comLibraryOpen sourcelatest reaction

  4. disco_rl

    Accompanying code for "Discovering State-of-the-art Reinforcement Algorithms" Nature publication

    GitHub750 starsgithub.comOpen sourcelatest reaction

  5. self-rewarding-lm-pytorch

    Implementation of the training framework proposed in Self-Rewarding Language Model, from MetaAI

    Hacker News145 ptsGitHub1,412 starsgithub.comLibraryOpen source2 reactionslatest reaction