SOTAVerified

Video Question Answering

Papers

Showing 110 of 460 papers

TitleStatusHype
Decoupled Seg Tokens Make Stronger Reasoning Video Segmenter and GrounderCode1
LLaVA-Scissor: Token Compression with Semantic Connected Components for Video LLMsCode2
How Far Can Off-the-Shelf Multimodal Large Language Models Go in Online Episodic Memory Question Answering?0
video-SALMONN 2: Captioning-Enhanced Audio-Visual Large Language ModelsCode2
CogStream: Context-guided Streaming Video Question Answering0
CausalVQA: A Physically Grounded Causal Reasoning Benchmark for Video ModelsCode2
V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and PlanningCode7
Looking Beyond Visible Cues: Implicit Video Question Answering via Dual-Clue ReasoningCode0
EgoVLM: Policy Optimization for Egocentric Video UnderstandingCode0
VUDG: A Dataset for Video Understanding Domain Generalization0
Show:102550
← PrevPage 1 of 46Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1VLAP (4 frames)Average Accuracy67.1Unverified
2LLaMA-VQAAverage Accuracy65.4Unverified
3SeViLAAverage Accuracy64.9Unverified
4InternVideoAverage Accuracy58.7Unverified
5GF(sup)Average Accuracy53.94Unverified
6GF(uns)Average Accuracy53.86Unverified
7MISTAverage Accuracy51.13Unverified
8Temp[ATP]Average Accuracy48.37Unverified
9AnyMAL-70B (0-shot)Average Accuracy48.2Unverified
10All-in-oneAverage Accuracy47.5Unverified