Embodied Question Answering on OpenEQA v1.0 (test)
67.7LLM-MatchCoV
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CoVBackbone=GLM-4.6v, Action Steps=62026.01 | 67.7 | 8.5 | 1.04 | |
| CoVBackbone=GLM-4.6v, Action Steps=72026.01 | 67.64 | — | — | |
| CoVBackbone=GLM-4.6v, Action Steps=22026.01 | 67.33 | — | — | |
| CoVBackbone=GLM-4.6v, Action Steps=52026.01 | 67.23 | — | — | |
| CoVBackbone=GLM-4.6v, Action Steps=32026.01 | 67.2 | — | — | |
| CoVBackbone=GLM-4.6v, Action Steps=12026.01 | 67 | — | — | |
| CoVBackbone=GLM-4.6v, Action Steps=42026.01 | 66.3 | — | — | |
| BaselineBackbone=GLM-4.6v2026.01 | 62.4 | — | — | |
| CoVBackbone=Qwen3-VL-Flash, Action Steps=42026.01 | 59.82 | 13.62 | 1.82 | |
| CoVBackbone=Qwen3-VL-Flash, Action Steps=72026.01 | 59.77 | — | — | |
| CoVBackbone=Qwen3-VL-Flash, Action Steps=62026.01 | 59.39 | — | — | |
| CoVBackbone=Gemini-2.5-Flash, Action Steps=62026.01 | 59.23 | 11.7 | 3.73 | |
| CoVBackbone=Gemini-2.5-Flash, Action Steps=72026.01 | 59.16 | — | — | |
| CoVBackbone=Qwen3-VL-Flash, Action Steps=22026.01 | 59.07 | — | — | |
| CoVBackbone=Qwen3-VL-Flash, Action Steps=32026.01 | 59.03 | — | — | |
| CoVBackbone=Qwen3-VL-Flash, Action Steps=52026.01 | 59.03 | — | — | |
| CoVBackbone=Gemini-2.5-Flash, Action Steps=52026.01 | 58.82 | — | — | |
| CoVBackbone=Gemini-2.5-Flash, Action Steps=32026.01 | 58.8 | — | — | |
| CoVBackbone=Qwen3-VL-Flash, Action Steps=12026.01 | 58.75 | — | — | |
| CoVBackbone=Gemini-2.5-Flash, Action Steps=42026.01 | 58.15 | — | — | |
| CoVBackbone=Gemini-2.5-Flash, Action Steps=22026.01 | 57.15 | — | — | |
| CoVBackbone=Gemini-2.5-Flash, Action Steps=12026.01 | 57.1 | — | — | |
| BaselineBackbone=Qwen3-VL-Flash2026.01 | 52.65 | — | — | |
| BaselineBackbone=Gemini-2.5-Flash2026.01 | 52.3 | — | — | |
| CoVBackbone=GPT-4o-Mini, Action Steps=72026.01 | 51.62 | — | — | |
| CoVBackbone=GPT-4o-Mini, Action Steps=52026.01 | 51.56 | 12.4 | 3.43 | |
| CoVBackbone=GPT-4o-Mini, Action Steps=62026.01 | 51.41 | — | — | |
| CoVBackbone=GPT-4o-Mini, Action Steps=42026.01 | 50.8 | — | — | |
| CoVBackbone=GPT-4o-Mini, Action Steps=22026.01 | 50.15 | — | — | |
| CoVBackbone=GPT-4o-Mini, Action Steps=12026.01 | 49.85 | — | — | |
| CoVBackbone=GPT-4o-Mini, Action Steps=32026.01 | 49.6 | — | — | |
| BaselineBackbone=GPT-4o-Mini2026.01 | 45.87 | — | — |