List your company

Math Reasoning on AIME 25 (pass@1)

40Pass@1OpenThinker3-7B
-1.69.22030.8Mar 1, 2026Mar 5, 2026Mar 9, 2026Mar 13, 2026Mar 17, 2026Mar 21, 2026Mar 26, 2026

Evaluation results

MethodLinks
OpenThinker3-7B2026.0340
teacher top-K local support matchingmasking=false2026.0326.7
teacher top-K local support matchingmasking=true2026.0323.3
Sampled-token OPDmasking=false2026.0316.7
Sampled-token OPDmasking=true2026.0316.7
IPPODebate Setting=Heterogeneous MAD, Agent=A1, Round (T)=5, Backbone=Qwen3-4B-Instruct-25072026.0313.3
UMADDebate Setting=Heterogeneous MAD, Agent=A1, Round (T)=2, Backbone=Qwen3-4B-Instruct-25072026.0313.3
UMADDebate Setting=Heterogeneous MAD, Agent=A1, Round (T)=5, Backbone=Qwen3-4B-Instruct-25072026.0313.3
IPPODebate Setting=Heterogeneous MAD, Agent=A1, Round (T)=2, Backbone=Qwen3-4B-Instruct-25072026.036.7
UMADDebate Setting=Heterogeneous MAD, Agent=A1, Round (T)=1, Backbone=Qwen3-4B-Instruct-25072026.036.7
Zero-ShotDebate Setting=Homogeneous MAD, Agent=A0, Round (T)=1, Backbone=Qwen2.5-3B-Instruct2026.030
Qwen2.5-7B-It2026.030