Spatial-Temporal Reasoning on MMSI
41.8AccuracyGPT-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-52026.02 | 41.8 | |
| Gemini-2.5-Pro2026.02 | 36.9 | |
| ReMoT-4b-CoTModel Size=4b, Thinking Mode=CoT2026.02 | 30.8 | |
| GPT-4o2026.02 | 30.3 | |
| InternVL2.5-8BModel Size=8B2026.02 | 28.7 | |
| Qwen3-VL-30B-CoTModel Size=30B, Thinking Mode=CoT2026.02 | 28.5 | |
| Qwen3-VL-4B-CoTModel Size=4B, Thinking Mode=CoT2026.02 | 26.8 | |
| Qwen2.5-VL-7BModel Size=7B2026.02 | 25.9 | |
| LLaVA-Next-7BModel Size=7B2026.02 | 24.5 |