List your company

Hallucination Reasoning on SugarCrepe

86.4AccuracyArgos
79.6481.39583.1584.905Dec 3, 2025

Evaluation results

MethodLinks
Argos2025.1286.4
Qwen2.5VL-7B2025.1285.2
Video-R1Training=SFT2025.1283.3
Qwen2.5VL-7BChain-of-Thought (CoT)=true2025.1283.2
Video-R1Training=RL2025.1279.9