Hallucination Reasoning on CounterCurate
85.3AccuracyArgos
Evaluation results
| Method | Links | |
|---|---|---|
| Argos2025.12 | 85.3 | |
| Video-R1Training=RL2025.12 | 63.6 | |
| Qwen2.5VL-7B2025.12 | 61.4 | |
| Video-R1Training=SFT2025.12 | 60.6 | |
| Qwen2.5VL-7BChain-of-Thought (CoT)=true2025.12 | 60.4 |