Video Based Generative Performance 3
評価指標
gpt-score
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 | gpt-score |
|---|---|
| chat-univi-unified-visual-representation | 3.46 |
| ts-llava-constructing-visual-tokens-through | 3.86 |
| videogpt-integrating-image-and-video-encoders | 3.74 |
| mvbench-a-comprehensive-multi-modal-video | 3.64 |
| pllava-parameter-free-llava-extension-from-1 | 3.9 |
| video-llama-an-instruction-tuned-audio-visual | 2.16 |
| video-chatgpt-towards-detailed-video | 2.62 |
| llama-adapter-v2-parameter-efficient-visual | 2.30 |
| vtimellm-empower-llm-to-grasp-video-moments | 3.40 |
| slowfast-llava-a-strong-training-free | 3.84 |
| st-llm-large-language-models-are-effective-1 | 3.74 |
| videochat-chat-centric-video-understanding | 2.53 |
| one-for-all-video-conversation-is-feasible | 3.27 |
| minigpt4-video-advancing-multimodal-llms-for | 3.57 |
| moviechat-from-dense-token-to-sparse-memory | 3.01 |
| ppllava-varied-video-sequence-understanding | 4.21 |
| mvbench-a-comprehensive-multi-modal-video | 3.51 |
| one-for-all-video-conversation-is-feasible | 2.89 |