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

Discriminative Reasoning on RPEval Multi-MACRO, Explicit Memory

0.4IA Score

Qwen2.5-7B + RP-Reasoner

Updated 1mo ago

Evaluation Results

Method	Links
Qwen2.5-7B + RP-Reasoner 2026.01		0.4	0.01	0.17
DeepSeek-V3 + CoT 2026.01		0.4	0.1	0.23
GPT-4.1 + RP-Reasoner 2026.01		0.38	0.2	0.27
GPT-5 + RP-Reasoner 2026.01		0.38	0.23	0.3
DeepSeek-V3 + RP-Reasoner 2026.01		0.35	0.29	0.31
GPT-4.1 + CoT 2026.01		0.2	0.09	0.13
Qwen2.5-7B 2026.01		0.18	0.01	0.08
Qwen2.5-7B + Reminder 2026.01		0.12	0.02	0.06
GPT-5 + CoT 2026.01		0.12	0.11	0.11
DeepSeek-V3 2026.01		0.08	0.04	0.06
GPT-4.1 + Reminder 2026.01		0.08	0.04	0.06
DeepSeek-V3 + Reminder 2026.01		0.05	0.07	0.06
GPT-4.1 2026.01		0.05	0.01	0.03
Qwen2.5-7B + CoT 2026.01		0.04	0.03	0.03
GPT-5 2026.01		0	0.04	0.03
GPT-5 + Reminder 2026.01		0	0.03	0.02