| video-based-generative-performance | IG-VLM-GPT4v | #8 | mean: 3.17Correctness of Information: 3.40Detail Orientation: 2.80… |
| zero-shot-video-question-answer-on-intentqa | IG-VLM | #5 | Accuracy: 65.3 |
| zero-shot-video-question-answer-on-next-qa | IG-VLM(LLaVA v1.6) | #8 | Accuracy: 70.9 |
| zero-shot-video-question-answer-on-next-qa | IG-VLM (GPT-4) | #10 | Accuracy: 68.6 |
| zero-shot-video-question-answer-on-tvqa | IG-VLM (no speech, GPT-4V) | #2 | Accuracy: 57.8 |
| zeroshot-video-question-answer-on-activitynet | IG-VLM | #7 | Accuracy: 58.4Confidence Score: 3.5 |
| zeroshot-video-question-answer-on-msrvtt-qa | IG-VLM | #9 | Accuracy: 63.8Confidence Score: 3.5 |
| zeroshot-video-question-answer-on-msvd-qa | IG-VLM-34B | #7 | Accuracy: 79.6Confidence Score: 4.1 |
| zeroshot-video-question-answer-on-tgif-qa | IG-VLM | #6 | Accuracy: 79.1Confidence Score: 4.2 |