reward-model 开源项目推荐（121 个）

agentscope-ai /OpenJudge

OpenJudge：整体评估与质量奖励的统一框架

wendell0218 /Awesome-RL-for-Video-Generation

A curated list of papers on reinforcement learning for video generation

opendilab /LightRFT

LightRFT: Light, Efficient, Omni-modal & Reward-model Driven Reinforcement Fine-Tuning Framework

alphadl /AdaRubrics

AdaRubric: Adaptive Dynamic Rubric Evaluator for Agent Trajectories

Tencent-Hunyuan /ArtifactsBenchmark

该项目在 GitHub 上没有填写描述

xlang-ai /CUA-Gym

Scalable pipeline for synthesizing verifiable RLVR training data for computer-use agents

DaoyuanLi2816 /pairjudge

Pairwise LLM judges (A/B/tie): budget-aware multi-turn packing, position-bias correction, pseudo-label distillation. Generalized from the 4th-place (gold) solution to Kaggle LMSYS Chatbot Arena.

VectorInstitute /vector-inference

Efficient LLM inference on Slurm clusters.

TIGER-AI-Lab /RewardHarness

Self-evolving agentic reward framework for image-editing evaluation — 47.4% on EditReward-Bench from only 100 preference demos, no reward-model training. arXiv 2605.08703.

Westlake-AI /SemiReward

[ICLR 2024] SemiReward: A General Reward Model for Semi-supervised Learning

bobxwu /learning-from-rewards-llm-papers

A comrephensive collection of learning from rewards in the post-training and test-time scaling of LLMs, with a focus on both reward models and learning strategies across training, inference, and post-inference stages.

InternLM /Visual-ERM

[NeurIPS 2026] Official Implementation of "Visual-ERM: Reward Modeling for Visual Equivalence"

tongjingqi /Awesome-Agent-RL

A curated list of awesome resources about reward construction for AI agents. This repository covers cutting-edge research, and practical guides on defining and collecting rewards to build more intelligent and aligned AI agents.

ZhangqiJiang07 /GEditBench_v2

GEditBench v2: A Human-Aligned Benchmark for General Image Editing

IgniaraosForge /bookish-octo-broccoli-respocket

Any of my divice

LARK-AI-Lab /CodeScaler

The official repo for "CodeScaler: Scaling Code LLM Training and Test-Time Inference via Execution-Free Reward Models"

Amirhosein-gh98 /Gnosis

Can LLMs Predict Their Own Failures? Self-Awareness via Internal Circuits

PlantPotatoOnMoon /HPSv3-PlusPlus

[ACM MM 2026] HPSv3++: Scaling Reward Models Across the Full Spectrum of Diffusion Model Capabilities

Qwen-Applications /OpenRS

Open Rubric System: Scaling Reinforcement Learning with Pairwise Adaptive Rubric

InternLM /Spark

An official implementation of "SPARK: Synergistic Policy And Reward Co-Evolving Framework"