List your company

Logical Reasoning on ProofWriter (Acc, TNFT, Tokens, Delay)

92.2AccVanilla
88.66489.58290.591.418May 12, 2026

Evaluation results

MethodLinks
VanillaBackbone=Qwen3-4B2026.0592.2195.141,125.1331.6
OursBackbone=Qwen3-4B2026.0591.80736.8424.56
VanillaBackbone=Qwen3-1.7B2026.0590.7195.141,227.1926.46
BaseBackbone=Qwen3-4B2026.0590.5197.142,066.162.94
OursBackbone=Qwen3-1.7B2026.0590.20714.7920.42
BaseBackbone=Qwen3-1.7B2026.0588.8197.142,110.2449.62