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.

benchmark

View on GitHub Website

92 Stars

9 Forks

1 Watchers

Python Language

other License

100 SrcLog Score

Cost to Build

$27.7K

Market Value

$78.5K

How is this calculated?

Growth over time

2 data points · 2026-10-03 → 2026-10-27

Stars Forks Watchers

💬

How do you feel about this project?

Ask AI about RewardHarness

Question copied to clipboard

What is the TIGER-AI-Lab/RewardHarness GitHub project? Description: "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.". Written in Python. Explain what it does, its main use cases, key features, and who would benefit from using it.

Question is copied to clipboard — paste it after the AI opens.

How to clone RewardHarness

Clone via HTTPS

git clone https://github.com/TIGER-AI-Lab/RewardHarness.git

Clone via SSH

[email protected]:TIGER-AI-Lab/RewardHarness.git

Download ZIP

Download main.zip

Found an issue?

Report bugs or request features on the RewardHarness issue tracker:

Open GitHub Issues

Similar to RewardHarness

netdata fashion-mnist FrameworkBenchmarks BenchmarkDotNet jmeter awesome-semantic-segmentation sysbench hyperfine tsung benchmark_results across web-frameworks php-framework-benchmark jsperf.com go-web-framework-benchmark huststore phoronix-test-suite Attabench ann-benchmarks sbt-jmh caffenet-benchmark chillout IocPerformance prophiler TBCF NBench sympact awesome-http-benchmark BlurTestAndroid pytest-benchmark