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

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

0.8958SRCC

RE-Reward

Updated 23d ago

Evaluation Results

Method	Links
RE-Reward 2026.05		0.8958	0.7315	0.9027
Gemini-3-Flash 2026.05		0.8438	0.7136	0.8173
ChatGPT-5.4 2026.05		0.7539	0.6528	0.7572
Gemma-4-31B 2026.05		0.7035	0.5902	0.7177
Qwen-3.5-35B-A3B 2026.05		0.5728	0.486	0.5956
Qwen3-VL-8B 2026.05		0.5386	0.4545	0.5913
Qwen-3.5-9B 2026.05		0.4097	0.3463	0.4787
mPLUG-Owl3-7B 2026.05		0.3473	0.2819	0.3197
FLEUR 2026.05		0.3384	0.2496	0.3258
DeepSeek-VL2-Small 2026.05		0.3318	0.2593	0.2843
MiniCPM-V-2.6-8B 2026.05		0.2749	0.2285	0.2269
InternVL3-8B 2026.05		0.1697	0.144	0.148
InternVL3.5-8B 2026.05		0.0784	0.0612	0.1798
CLIP-S-ViT-B/32 2026.05		0.034	0.0238	0.0417
Ovis2.5-9B 2026.05		0.0314	0.0268	0.0419
PAC-S++ 2026.05		0.0298	0.0209	0.0148