Video Reasoning on VidHalluc (test)
81.15Binary QA Accuracy (ACH)GPT-4o
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-4oLLM Params=~200B2026.02 | 81.15 | 90.95 | 82 | 71.58 | |
| Gemini-1.5-ProLLM Params=~200B2026.02 | 75.27 | 79.25 | 83.83 | 63.96 | |
| GraphThinkerLLM Params=7B, EVSG=true, RL=true, finetuned_on_Rextime=true2026.02 | 66.04 | 84.57 | 76.33 | 57.81 | |
| VILA1.5LLM Params=13B2026.02 | 58.46 | 81.88 | 63.33 | 35.03 | |
| QwenVL-2.5+EVSGLLM Params=7B, EVSG=true2026.02 | 50.92 | 83.23 | 75.33 | 54.58 | |
| QwenVL-2.5LLM Params=7B2026.02 | 50.46 | 81.99 | 67.5 | 46.77 | |
| VideoLLaMA2LLM Params=7B2026.02 | 50.04 | 83.84 | 26.17 | 65.12 | |
| PLLaVALLM Params=13B2026.02 | 35.3 | 76.96 | 16.5 | 32.44 | |
| ShareGPT4VideoLLM Params=8B2026.02 | 29.96 | 44.78 | 49.5 | 17.83 | |
| Video-LLAVALLM Params=7B2026.02 | 26.84 | 64.45 | 27.17 | 29.6 | |
| LLaVA-NeXT-VideoLLM Params=34B2026.02 | 26.6 | 77.57 | 21.33 | 44.4 | |
| Chat-UniViLLM Params=13B2026.02 | 23.77 | 54.79 | 35.5 | 29.87 | |
| Video-ChatGPTLLM Params=7B2026.02 | 9.5 | 24.58 | 30.17 | 7.7 |