Share your thoughts, 1 month free Claude Pro on usSee more

Reward Modeling (Accuracy) on ReasonEdit-Reward-113K 1.0 (test)

0.9323SRCC

RE-Reward

Updated 23d ago

Evaluation Results

Method	Links
RE-Reward 2026.05		0.9323	0.7781	0.9316
Gemini-3-Flash 2026.05		0.8587	0.7209	0.8411
ChatGPT-5.4 2026.05		0.8532	0.7284	0.8535
Gemma-4-31B 2026.05		0.8264	0.6943	0.8251
Qwen-3.5-35B-A3B 2026.05		0.7761	0.6598	0.7797
FLEUR 2026.05		0.4853	0.3612	0.5055
DeepSeek-VL2-Small 2026.05		0.4462	0.3525	0.3839
Qwen-3.5-9B 2026.05		0.4279	0.3544	0.432
Qwen3-VL-8B 2026.05		0.3947	0.3266	0.4175
mPLUG-Owl3-7B 2026.05		0.3835	0.3163	0.3894
MiniCPM-V-2.6-8B 2026.05		0.1548	0.1287	0.1538
Ovis2.5-9B 2026.05		0.0752	0.0725	0.0453
InternVL3-8B 2026.05		0.0687	0.0553	0.0929
CLIP-S-ViT-B/32 2026.05		0.0655	0.0452	0.0624
InternVL3.5-8B 2026.05		0.0497	0.041	0.043
PAC-S++ 2026.05		0.005	0.0029	0.0062