{"task":"Zero-Shot Video Question Answer","dataset":"MSRVTT-QA","metric_names":["Accuracy","Confidence Score"],"rows":[{"id":24778,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Flash-VStream","metrics":{"Accuracy":"72.4","Confidence Score":"3.4"},"paper_url":"https://arxiv.org/abs/2406.08085v2","paper_title":"Flash-VStream: Memory-Based Real-Time Understanding for Long Video Streams","paper_date":"2024-06-12","code_links":[{"title":"IVGSZ/Flash-VStream","url":"https://github.com/IVGSZ/Flash-VStream"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24779,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"PLLaVA (34B)","metrics":{"Accuracy":"68.7","Confidence Score":"3.6"},"paper_url":"https://arxiv.org/abs/2404.16994v2","paper_title":"PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning","paper_date":"2024-04-25","code_links":[{"title":"magic-research/PLLaVA","url":"https://github.com/magic-research/PLLaVA"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24780,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Elysium","metrics":{"Accuracy":"67.5","Confidence Score":"3.2"},"paper_url":"https://arxiv.org/abs/2403.16558v2","paper_title":"Elysium: Exploring Object-level Perception in Videos via MLLM","paper_date":"2024-03-25","code_links":[{"title":"hon-wong/elysium","url":"https://github.com/hon-wong/elysium"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24781,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"SlowFast-LLaVA-34B","metrics":{"Accuracy":"67.4","Confidence Score":"3.7"},"paper_url":"https://arxiv.org/abs/2407.15841v2","paper_title":"SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models","paper_date":"2024-07-22","code_links":[{"title":"apple/ml-slowfast-llava","url":"https://github.com/apple/ml-slowfast-llava"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24782,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Tarsier (34B)","metrics":{"Accuracy":"66.4","Confidence Score":"3.7"},"paper_url":"https://arxiv.org/abs/2407.00634v2","paper_title":"Tarsier: Recipes for Training and Evaluating Large Video Description Models","paper_date":"2024-06-30","code_links":[{"title":"bytedance/tarsier","url":"https://github.com/bytedance/tarsier"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24783,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"LinVT-Qwen2-VL\n(7B)","metrics":{"Accuracy":"66.2","Confidence Score":"4.0"},"paper_url":"https://arxiv.org/abs/2412.05185v2","paper_title":"LinVT: Empower Your Image-level Large Language Model to Understand Videos","paper_date":"2024-12-06","code_links":[{"title":"gls0425/linvt","url":"https://github.com/gls0425/linvt"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24784,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"TS-LLaVA-34B","metrics":{"Accuracy":"66.2","Confidence Score":"3.6"},"paper_url":"https://arxiv.org/abs/2411.11066v1","paper_title":"TS-LLaVA: Constructing Visual Tokens through Thumbnail-and-Sampling for Training-Free Video Large Language Models","paper_date":"2024-11-17","code_links":[{"title":"tingyu215/ts-llava","url":"https://github.com/tingyu215/ts-llava"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24785,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"PPLLaVA-7B","metrics":{"Accuracy":"64.3","Confidence Score":"3.5"},"paper_url":"https://arxiv.org/abs/2411.02327v2","paper_title":"PPLLaVA: Varied Video Sequence Understanding With Prompt Guidance","paper_date":"2024-11-04","code_links":[{"title":"farewellthree/ppllava","url":"https://github.com/farewellthree/ppllava"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24786,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"IG-VLM","metrics":{"Accuracy":"63.8","Confidence Score":"3.5"},"paper_url":"https://arxiv.org/abs/2403.18406v1","paper_title":"An Image Grid Can Be Worth a Video: Zero-shot Video Question Answering Using a VLM","paper_date":"2024-03-27","code_links":[{"title":"imagegridworth/IG-VLM","url":"https://github.com/imagegridworth/IG-VLM"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24787,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"ST-LLM","metrics":{"Accuracy":"63.2","Confidence Score":"3.4"},"paper_url":"https://arxiv.org/abs/2404.00308v1","paper_title":"ST-LLM: Large Language Models Are Effective Temporal Learners","paper_date":"2024-03-30","code_links":[{"title":"TencentARC/ST-LLM","url":"https://github.com/TencentARC/ST-LLM"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24788,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"CAT-7B","metrics":{"Accuracy":"62.1","Confidence Score":"3.5"},"paper_url":"https://arxiv.org/abs/2403.04640v1","paper_title":"CAT: Enhancing Multimodal Large Language Model to Answer Questions in Dynamic Audio-Visual Scenarios","paper_date":"2024-03-07","code_links":[{"title":"rikeilong/bay-cat","url":"https://github.com/rikeilong/bay-cat"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24789,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"VideoGPT+","metrics":{"Accuracy":"60.6","Confidence Score":"3.6"},"paper_url":"https://arxiv.org/abs/2406.09418v1","paper_title":"VideoGPT+: Integrating Image and Video Encoders for Enhanced Video Understanding","paper_date":"2024-06-13","code_links":[{"title":"mbzuai-oryx/videogpt-plus","url":"https://github.com/mbzuai-oryx/videogpt-plus"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24790,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Vista-LLaMA-7B","metrics":{"Accuracy":"60.5","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2312.08870v1","paper_title":"Vista-LLaMA: Reliable Video Narrator via Equal Distance to Visual Tokens","paper_date":"2023-12-12","code_links":[],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24791,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"MiniGPT4-video-7B","metrics":{"Accuracy":"59.73"},"paper_url":"https://arxiv.org/abs/2404.03413v1","paper_title":"MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens","paper_date":"2024-04-04","code_links":[{"title":"Vision-CAIR/MiniGPT4-video","url":"https://github.com/Vision-CAIR/MiniGPT4-video"},{"title":"pwc-1/Paper-9","url":"https://github.com/pwc-1/Paper-9/tree/main/2/minigpt4"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24792,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"LLaVA-Mini","metrics":{"Accuracy":"59.5","Confidence Score":"3.6"},"paper_url":"https://arxiv.org/abs/2501.03895v1","paper_title":"LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token","paper_date":"2025-01-07","code_links":[{"title":"ictnlp/llava-mini","url":"https://github.com/ictnlp/llava-mini"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24793,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Video-LaVIT","metrics":{"Accuracy":"59.3","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2402.03161v3","paper_title":"Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization","paper_date":"2024-02-05","code_links":[{"title":"jy0205/lavit","url":"https://github.com/jy0205/lavit"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24794,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Video-LLaVA-7B","metrics":{"Accuracy":"59.2","Confidence Score":"3.5"},"paper_url":"https://arxiv.org/abs/2311.10122v3","paper_title":"Video-LLaVA: Learning United Visual Representation by Alignment Before Projection","paper_date":"2023-11-16","code_links":[{"title":"PKU-YuanGroup/Video-LLaVA","url":"https://github.com/PKU-YuanGroup/Video-LLaVA"},{"title":"PKU-YuanGroup/MoE-LLaVA","url":"https://github.com/PKU-YuanGroup/MoE-LLaVA"},{"title":"pku-yuangroup/languagebind","url":"https://github.com/pku-yuangroup/languagebind"},{"title":"pku-yuangroup/video-bench","url":"https://github.com/pku-yuangroup/video-bench"},{"title":"qiujihao19/artemis","url":"https://github.com/qiujihao19/artemis"},{"title":"PKU-YuanGroup/LLMBind","url":"https://github.com/PKU-YuanGroup/LLMBind"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":1,"source":"archive","tags":[]},{"id":24795,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"LLaMA-VID-13B (2 Token)","metrics":{"Accuracy":"58.9","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2311.17043v1","paper_title":"LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models","paper_date":"2023-11-28","code_links":[{"title":"lastmile-ai/aiconfig","url":"https://github.com/lastmile-ai/aiconfig"},{"title":"dvlab-research/llama-vid","url":"https://github.com/dvlab-research/llama-vid"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24796,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"LLaMA-VID-7B (2 Token)","metrics":{"Accuracy":"57.7","Confidence Score":"3.2"},"paper_url":"https://arxiv.org/abs/2311.17043v1","paper_title":"LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models","paper_date":"2023-11-28","code_links":[{"title":"lastmile-ai/aiconfig","url":"https://github.com/lastmile-ai/aiconfig"},{"title":"dvlab-research/llama-vid","url":"https://github.com/dvlab-research/llama-vid"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24797,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"SUM-shot+Vicuna","metrics":{"Accuracy":"56.8"},"paper_url":"https://arxiv.org/abs/2312.10300v2","paper_title":"Shot2Story20K: A New Benchmark for Comprehensive Understanding of Multi-shot Videos","paper_date":"2023-12-16","code_links":[{"title":"bytedance/Shot2Story","url":"https://github.com/bytedance/Shot2Story"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24798,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Omni-VideoAssistant","metrics":{"Accuracy":"55.3","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2308.04126v2","paper_title":"OmniDataComposer: A Unified Data Structure for Multimodal Data Fusion and Infinite Data Generation","paper_date":"2023-08-08","code_links":[{"title":"shajiayu1/OmniDataComposer","url":"https://github.com/shajiayu1/OmniDataComposer"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24799,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Chat-UniVi-7B","metrics":{"Accuracy":"55.0","Confidence Score":"3.1"},"paper_url":"https://arxiv.org/abs/2311.08046v3","paper_title":"Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding","paper_date":"2023-11-14","code_links":[{"title":"pku-yuangroup/chat-univi","url":"https://github.com/pku-yuangroup/chat-univi"},{"title":"skyworkai/moh","url":"https://github.com/skyworkai/moh"},{"title":"skyworkai/moe-plus-plus","url":"https://github.com/skyworkai/moe-plus-plus"},{"title":"pku-yuangroup/video-bench","url":"https://github.com/pku-yuangroup/video-bench"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24800,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"VideoChat2","metrics":{"Accuracy":"54.1","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2311.17005v4","paper_title":"MVBench: A Comprehensive Multi-modal Video Understanding Benchmark","paper_date":"2023-11-28","code_links":[{"title":"opengvlab/ask-anything","url":"https://github.com/opengvlab/ask-anything"},{"title":"magic-research/PLLaVA","url":"https://github.com/magic-research/PLLaVA"},{"title":"bytedance/tarsier","url":"https://github.com/bytedance/tarsier"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24801,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"MovieChat","metrics":{"Accuracy":"52.7","Confidence Score":"2.6"},"paper_url":"https://arxiv.org/abs/2307.16449v4","paper_title":"MovieChat: From Dense Token to Sparse Memory for Long Video Understanding","paper_date":"2023-07-31","code_links":[{"title":"rese1f/MovieChat","url":"https://github.com/rese1f/MovieChat"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24802,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"BT-Adapter (zero-shot)","metrics":{"Accuracy":"51.2","Confidence Score":"2.9"},"paper_url":"https://arxiv.org/abs/2309.15785v2","paper_title":"BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning","paper_date":"2023-09-27","code_links":[{"title":"farewellthree/BT-Adapter","url":"https://github.com/farewellthree/BT-Adapter"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24803,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"BT-Adapter (zero-shot)","metrics":{"Accuracy":"51.2","Confidence Score":"2.9"},"paper_url":"https://arxiv.org/abs/2309.15785v2","paper_title":"BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning","paper_date":"2023-09-27","code_links":[{"title":"farewellthree/BT-Adapter","url":"https://github.com/farewellthree/BT-Adapter"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24804,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Video-ChatGPT-7B","metrics":{"Accuracy":"49.3","Confidence Score":"2.8"},"paper_url":"https://arxiv.org/abs/2306.05424v2","paper_title":"Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models","paper_date":"2023-06-08","code_links":[{"title":"mbzuai-oryx/video-chatgpt","url":"https://github.com/mbzuai-oryx/video-chatgpt"},{"title":"qiujihao19/artemis","url":"https://github.com/qiujihao19/artemis"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24805,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Video Chat-7B","metrics":{"Accuracy":"45.0","Confidence Score":"2.5"},"paper_url":"https://arxiv.org/abs/2305.06355v2","paper_title":"VideoChat: Chat-Centric Video Understanding","paper_date":"2023-05-10","code_links":[{"title":"opengvlab/ask-anything","url":"https://github.com/opengvlab/ask-anything"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24806,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"LLaMA Adapter-7B","metrics":{"Accuracy":"43.8","Confidence Score":"2.7"},"paper_url":"https://arxiv.org/abs/2304.15010v1","paper_title":"LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model","paper_date":"2023-04-28","code_links":[{"title":"opengvlab/llama-adapter","url":"https://github.com/opengvlab/llama-adapter"},{"title":"zrrskywalker/llama-adapter","url":"https://github.com/zrrskywalker/llama-adapter"},{"title":"Mind23-2/MindCode-140","url":"https://github.com/Mind23-2/MindCode-140"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":24807,"task":"Zero-Shot Video Question Answer","parent_task":"Question Answering","dataset":"MSRVTT-QA","model_name":"Video LLaMA-7B","metrics":{"Accuracy":"29.6","Confidence Score":"1.8"},"paper_url":"https://arxiv.org/abs/2306.02858v4","paper_title":"Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding","paper_date":"2023-06-05","code_links":[{"title":"damo-nlp-sg/video-llama","url":"https://github.com/damo-nlp-sg/video-llama"},{"title":"damo-nlp-sg/videollama2","url":"https://github.com/damo-nlp-sg/videollama2"},{"title":"damo-nlp-sg/videollama3","url":"https://github.com/damo-nlp-sg/videollama3"},{"title":"xinding-sys/StreamMind","url":"https://github.com/xinding-sys/StreamMind"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":27129,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Flash-VStream","metrics":{"Accuracy":"72.4","Confidence Score":"3.4"},"paper_url":"https://arxiv.org/abs/2406.08085v2","paper_title":"Flash-VStream: Memory-Based Real-Time Understanding for Long Video Streams","paper_date":"2024-06-12","code_links":[{"title":"IVGSZ/Flash-VStream","url":"https://github.com/IVGSZ/Flash-VStream"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27130,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"PLLaVA (34B)","metrics":{"Accuracy":"68.7","Confidence Score":"3.6"},"paper_url":"https://arxiv.org/abs/2404.16994v2","paper_title":"PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning","paper_date":"2024-04-25","code_links":[{"title":"magic-research/PLLaVA","url":"https://github.com/magic-research/PLLaVA"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27131,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Elysium","metrics":{"Accuracy":"67.5","Confidence Score":"3.2"},"paper_url":"https://arxiv.org/abs/2403.16558v2","paper_title":"Elysium: Exploring Object-level Perception in Videos via MLLM","paper_date":"2024-03-25","code_links":[{"title":"hon-wong/elysium","url":"https://github.com/hon-wong/elysium"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27132,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"SlowFast-LLaVA-34B","metrics":{"Accuracy":"67.4","Confidence Score":"3.7"},"paper_url":"https://arxiv.org/abs/2407.15841v2","paper_title":"SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models","paper_date":"2024-07-22","code_links":[{"title":"apple/ml-slowfast-llava","url":"https://github.com/apple/ml-slowfast-llava"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27133,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Tarsier (34B)","metrics":{"Accuracy":"66.4","Confidence Score":"3.7"},"paper_url":"https://arxiv.org/abs/2407.00634v2","paper_title":"Tarsier: Recipes for Training and Evaluating Large Video Description Models","paper_date":"2024-06-30","code_links":[{"title":"bytedance/tarsier","url":"https://github.com/bytedance/tarsier"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27134,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"LinVT-Qwen2-VL\n(7B)","metrics":{"Accuracy":"66.2","Confidence Score":"4.0"},"paper_url":"https://arxiv.org/abs/2412.05185v2","paper_title":"LinVT: Empower Your Image-level Large Language Model to Understand Videos","paper_date":"2024-12-06","code_links":[{"title":"gls0425/linvt","url":"https://github.com/gls0425/linvt"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27135,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"TS-LLaVA-34B","metrics":{"Accuracy":"66.2","Confidence Score":"3.6"},"paper_url":"https://arxiv.org/abs/2411.11066v1","paper_title":"TS-LLaVA: Constructing Visual Tokens through Thumbnail-and-Sampling for Training-Free Video Large Language Models","paper_date":"2024-11-17","code_links":[{"title":"tingyu215/ts-llava","url":"https://github.com/tingyu215/ts-llava"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27136,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"PPLLaVA-7B","metrics":{"Accuracy":"64.3","Confidence Score":"3.5"},"paper_url":"https://arxiv.org/abs/2411.02327v2","paper_title":"PPLLaVA: Varied Video Sequence Understanding With Prompt Guidance","paper_date":"2024-11-04","code_links":[{"title":"farewellthree/ppllava","url":"https://github.com/farewellthree/ppllava"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27137,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"IG-VLM","metrics":{"Accuracy":"63.8","Confidence Score":"3.5"},"paper_url":"https://arxiv.org/abs/2403.18406v1","paper_title":"An Image Grid Can Be Worth a Video: Zero-shot Video Question Answering Using a VLM","paper_date":"2024-03-27","code_links":[{"title":"imagegridworth/IG-VLM","url":"https://github.com/imagegridworth/IG-VLM"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27138,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"ST-LLM","metrics":{"Accuracy":"63.2","Confidence Score":"3.4"},"paper_url":"https://arxiv.org/abs/2404.00308v1","paper_title":"ST-LLM: Large Language Models Are Effective Temporal Learners","paper_date":"2024-03-30","code_links":[{"title":"TencentARC/ST-LLM","url":"https://github.com/TencentARC/ST-LLM"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27139,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"CAT-7B","metrics":{"Accuracy":"62.1","Confidence Score":"3.5"},"paper_url":"https://arxiv.org/abs/2403.04640v1","paper_title":"CAT: Enhancing Multimodal Large Language Model to Answer Questions in Dynamic Audio-Visual Scenarios","paper_date":"2024-03-07","code_links":[{"title":"rikeilong/bay-cat","url":"https://github.com/rikeilong/bay-cat"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27140,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"VideoGPT+","metrics":{"Accuracy":"60.6","Confidence Score":"3.6"},"paper_url":"https://arxiv.org/abs/2406.09418v1","paper_title":"VideoGPT+: Integrating Image and Video Encoders for Enhanced Video Understanding","paper_date":"2024-06-13","code_links":[{"title":"mbzuai-oryx/videogpt-plus","url":"https://github.com/mbzuai-oryx/videogpt-plus"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27141,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Vista-LLaMA-7B","metrics":{"Accuracy":"60.5","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2312.08870v1","paper_title":"Vista-LLaMA: Reliable Video Narrator via Equal Distance to Visual Tokens","paper_date":"2023-12-12","code_links":[],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27142,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"MiniGPT4-video-7B","metrics":{"Accuracy":"59.73"},"paper_url":"https://arxiv.org/abs/2404.03413v1","paper_title":"MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens","paper_date":"2024-04-04","code_links":[{"title":"Vision-CAIR/MiniGPT4-video","url":"https://github.com/Vision-CAIR/MiniGPT4-video"},{"title":"pwc-1/Paper-9","url":"https://github.com/pwc-1/Paper-9/tree/main/2/minigpt4"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27143,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"LLaVA-Mini","metrics":{"Accuracy":"59.5","Confidence Score":"3.6"},"paper_url":"https://arxiv.org/abs/2501.03895v1","paper_title":"LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token","paper_date":"2025-01-07","code_links":[{"title":"ictnlp/llava-mini","url":"https://github.com/ictnlp/llava-mini"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27144,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Video-LaVIT","metrics":{"Accuracy":"59.3","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2402.03161v3","paper_title":"Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization","paper_date":"2024-02-05","code_links":[{"title":"jy0205/lavit","url":"https://github.com/jy0205/lavit"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27145,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Video-LLaVA-7B","metrics":{"Accuracy":"59.2","Confidence Score":"3.5"},"paper_url":"https://arxiv.org/abs/2311.10122v3","paper_title":"Video-LLaVA: Learning United Visual Representation by Alignment Before Projection","paper_date":"2023-11-16","code_links":[{"title":"PKU-YuanGroup/Video-LLaVA","url":"https://github.com/PKU-YuanGroup/Video-LLaVA"},{"title":"PKU-YuanGroup/MoE-LLaVA","url":"https://github.com/PKU-YuanGroup/MoE-LLaVA"},{"title":"pku-yuangroup/languagebind","url":"https://github.com/pku-yuangroup/languagebind"},{"title":"pku-yuangroup/video-bench","url":"https://github.com/pku-yuangroup/video-bench"},{"title":"qiujihao19/artemis","url":"https://github.com/qiujihao19/artemis"},{"title":"PKU-YuanGroup/LLMBind","url":"https://github.com/PKU-YuanGroup/LLMBind"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":1,"source":"archive","tags":[]},{"id":27146,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"LLaMA-VID-13B (2 Token)","metrics":{"Accuracy":"58.9","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2311.17043v1","paper_title":"LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models","paper_date":"2023-11-28","code_links":[{"title":"lastmile-ai/aiconfig","url":"https://github.com/lastmile-ai/aiconfig"},{"title":"dvlab-research/llama-vid","url":"https://github.com/dvlab-research/llama-vid"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27147,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"LLaMA-VID-7B (2 Token)","metrics":{"Accuracy":"57.7","Confidence Score":"3.2"},"paper_url":"https://arxiv.org/abs/2311.17043v1","paper_title":"LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models","paper_date":"2023-11-28","code_links":[{"title":"lastmile-ai/aiconfig","url":"https://github.com/lastmile-ai/aiconfig"},{"title":"dvlab-research/llama-vid","url":"https://github.com/dvlab-research/llama-vid"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27148,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"SUM-shot+Vicuna","metrics":{"Accuracy":"56.8"},"paper_url":"https://arxiv.org/abs/2312.10300v2","paper_title":"Shot2Story20K: A New Benchmark for Comprehensive Understanding of Multi-shot Videos","paper_date":"2023-12-16","code_links":[{"title":"bytedance/Shot2Story","url":"https://github.com/bytedance/Shot2Story"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27149,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Omni-VideoAssistant","metrics":{"Accuracy":"55.3","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2308.04126v2","paper_title":"OmniDataComposer: A Unified Data Structure for Multimodal Data Fusion and Infinite Data Generation","paper_date":"2023-08-08","code_links":[{"title":"shajiayu1/OmniDataComposer","url":"https://github.com/shajiayu1/OmniDataComposer"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27150,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Chat-UniVi-7B","metrics":{"Accuracy":"55.0","Confidence Score":"3.1"},"paper_url":"https://arxiv.org/abs/2311.08046v3","paper_title":"Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding","paper_date":"2023-11-14","code_links":[{"title":"pku-yuangroup/chat-univi","url":"https://github.com/pku-yuangroup/chat-univi"},{"title":"skyworkai/moh","url":"https://github.com/skyworkai/moh"},{"title":"skyworkai/moe-plus-plus","url":"https://github.com/skyworkai/moe-plus-plus"},{"title":"pku-yuangroup/video-bench","url":"https://github.com/pku-yuangroup/video-bench"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27151,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"VideoChat2","metrics":{"Accuracy":"54.1","Confidence Score":"3.3"},"paper_url":"https://arxiv.org/abs/2311.17005v4","paper_title":"MVBench: A Comprehensive Multi-modal Video Understanding Benchmark","paper_date":"2023-11-28","code_links":[{"title":"opengvlab/ask-anything","url":"https://github.com/opengvlab/ask-anything"},{"title":"magic-research/PLLaVA","url":"https://github.com/magic-research/PLLaVA"},{"title":"bytedance/tarsier","url":"https://github.com/bytedance/tarsier"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27152,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"MovieChat","metrics":{"Accuracy":"52.7","Confidence Score":"2.6"},"paper_url":"https://arxiv.org/abs/2307.16449v4","paper_title":"MovieChat: From Dense Token to Sparse Memory for Long Video Understanding","paper_date":"2023-07-31","code_links":[{"title":"rese1f/MovieChat","url":"https://github.com/rese1f/MovieChat"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27153,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"BT-Adapter (zero-shot)","metrics":{"Accuracy":"51.2","Confidence Score":"2.9"},"paper_url":"https://arxiv.org/abs/2309.15785v2","paper_title":"BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning","paper_date":"2023-09-27","code_links":[{"title":"farewellthree/BT-Adapter","url":"https://github.com/farewellthree/BT-Adapter"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27154,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"BT-Adapter (zero-shot)","metrics":{"Accuracy":"51.2","Confidence Score":"2.9"},"paper_url":"https://arxiv.org/abs/2309.15785v2","paper_title":"BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning","paper_date":"2023-09-27","code_links":[{"title":"farewellthree/BT-Adapter","url":"https://github.com/farewellthree/BT-Adapter"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27155,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Video-ChatGPT-7B","metrics":{"Accuracy":"49.3","Confidence Score":"2.8"},"paper_url":"https://arxiv.org/abs/2306.05424v2","paper_title":"Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models","paper_date":"2023-06-08","code_links":[{"title":"mbzuai-oryx/video-chatgpt","url":"https://github.com/mbzuai-oryx/video-chatgpt"},{"title":"qiujihao19/artemis","url":"https://github.com/qiujihao19/artemis"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27156,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Video Chat-7B","metrics":{"Accuracy":"45.0","Confidence Score":"2.5"},"paper_url":"https://arxiv.org/abs/2305.06355v2","paper_title":"VideoChat: Chat-Centric Video Understanding","paper_date":"2023-05-10","code_links":[{"title":"opengvlab/ask-anything","url":"https://github.com/opengvlab/ask-anything"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27157,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"LLaMA Adapter-7B","metrics":{"Accuracy":"43.8","Confidence Score":"2.7"},"paper_url":"https://arxiv.org/abs/2304.15010v1","paper_title":"LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model","paper_date":"2023-04-28","code_links":[{"title":"opengvlab/llama-adapter","url":"https://github.com/opengvlab/llama-adapter"},{"title":"zrrskywalker/llama-adapter","url":"https://github.com/zrrskywalker/llama-adapter"},{"title":"Mind23-2/MindCode-140","url":"https://github.com/Mind23-2/MindCode-140"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":27158,"task":"Zero-Shot Video Question Answer","parent_task":"Video Question Answering","dataset":"MSRVTT-QA","model_name":"Video LLaMA-7B","metrics":{"Accuracy":"29.6","Confidence Score":"1.8"},"paper_url":"https://arxiv.org/abs/2306.02858v4","paper_title":"Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding","paper_date":"2023-06-05","code_links":[{"title":"damo-nlp-sg/video-llama","url":"https://github.com/damo-nlp-sg/video-llama"},{"title":"damo-nlp-sg/videollama2","url":"https://github.com/damo-nlp-sg/videollama2"},{"title":"damo-nlp-sg/videollama3","url":"https://github.com/damo-nlp-sg/videollama3"},{"title":"xinding-sys/StreamMind","url":"https://github.com/xinding-sys/StreamMind"}],"metrics_order":"[\"Accuracy\", \"Confidence Score\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]}]}