Zeroshot Video Question Answer On Msvd Qa

评估指标

Accuracy

Confidence Score

评测结果

各个模型在此基准测试上的表现结果

模型名称	Accuracy	Confidence Score	Paper Title	Repository
BT-Adapter (zero-shot)	67.0	3.6	BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning	-
VideoGPT+	72.4	3.6	VideoGPT+: Integrating Image and Video Encoders for Enhanced Video Understanding	-
SlowFast-LLaVA-34B	79.9	4.1	SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models	-
Video-ChatGPT-7B	64.9	3.3	Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models	-
PPLLaVA-7B	77.1	4.0	PPLLaVA: Varied Video Sequence Understanding With Prompt Guidance	-
Elysium	75.8	3.7	Elysium: Exploring Object-level Perception in Videos via MLLM	-
BT-Adapter (zero-shot)	67.0	3.6	BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning	-
Video Chat-7B	56.3	2.8	VideoChat: Chat-Centric Video Understanding	-
Flash-VStream	80.3	3.9	Flash-VStream: Memory-Based Real-Time Understanding for Long Video Streams	-
VILA1.5-40B	80.1	-	VILA: On Pre-training for Visual Language Models	-
LLaVA-Mini	70.9	4.0	LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token	-
TS-LLaVA-34B	79.4	4.1	TS-LLaVA: Constructing Visual Tokens through Thumbnail-and-Sampling for Training-Free Video Large Language Models	-
IG-VLM-34B	79.6	4.1	An Image Grid Can Be Worth a Video: Zero-shot Video Question Answering Using a VLM	-
LLaMA-VID-7B (2 Token)	69.7	3.7	LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models	-
VideoChat2	70.0	3.9	MVBench: A Comprehensive Multi-modal Video Understanding Benchmark	-
Video LLaMA-7B	51.6	2.5	Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding	-
LinVT-Qwen2-VL (7B)	80.2	4.4	LinVT: Empower Your Image-level Large Language Model to Understand Videos	-
PLLaVA (34B)	79.9	4.2	PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning	-
LLaMA Adapter-7B	54.9	3.1	LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model	-
FrozenBiLM	33.8	-	Zero-Shot Video Question Answering via Frozen Bidirectional Language Models	-

0 of 28 row(s) selected.