Zero Shot Video Question Answer On Egoschema 1
評価指標
Accuracy
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | Accuracy |
|---|---|
| timechat-a-time-sensitive-multimodal-large | 33.0 |
| mvbench-a-comprehensive-multi-modal-video | 54.4 |
| tarsier-recipes-for-training-and-evaluating-1 | 61.7 |
| a-simple-llm-framework-for-long-range-video | 50.3 |
| internvideo-general-video-foundation-models | 32.1 |
| videotree-adaptive-tree-based-video | 61.1 |
| vamos-versatile-action-models-for-video | 53.6 |
| understanding-long-videos-in-one-multimodal | 37.6 |
| self-chained-image-language-model-for-video-1 | 22.7 |
| too-many-frames-not-all-useful-efficient | 61.1 |
| bimba-selective-scan-compression-for-long | 71.14 |
| モデル 12 | 20.0 |
| vamos-versatile-action-models-for-video | 48.3 |
| mvbench-a-comprehensive-multi-modal-video | 55.8 |
| videollama-2-advancing-spatial-temporal | 63.9 |
| video-rag-visually-aligned-retrieval | 66.7 |
| mplug-owl-modularization-empowers-large | 31.1 |
| linvt-empower-your-image-level-large-language | 69.5 |
| traveler-a-multi-lmm-agent-framework-for | 53.3 |
| mvbench-a-comprehensive-multi-modal-video | 56.7 |
| a-simple-llm-framework-for-long-range-video | 33.5 |
| video-recap-recursive-captioning-of-hour-long | 50.23 |
| zero-shot-video-question-answering-via-frozen | 26.9 |
| vamos-versatile-action-models-for-video | 36.7 |
| language-repository-for-long-video | 41.2 |
| internvideo2-scaling-video-foundation-models | 60.2 |
| longvu-spatiotemporal-adaptive-compression | 67.6 |