{"task":"Video-based Generative Performance Benchmarking","dataset":"VideoInstruct","metric_names":["mean","Correctness of Information","Detail Orientation","Contextual Understanding","Temporal Understanding","Consistency"],"rows":[{"id":26640,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"PPLLaVA-7B-dpo","metrics":{"Consistency":"3.81","Contextual Understanding":"4.21","Correctness of Information":"3.85","Detail Orientation":"3.56","Temporal Understanding":"3.21","mean":"3.73"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26641,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VLM-RLAIF","metrics":{"Consistency":"3.32","Contextual Understanding":"4","Correctness of Information":"3.63","Detail Orientation":"3.25","Temporal Understanding":"3.23","mean":"3.49"},"paper_url":"https://arxiv.org/abs/2402.03746v3","paper_title":"Tuning Large Multimodal Models for Videos using Reinforcement Learning from AI Feedback","paper_date":"2024-02-06","code_links":[{"title":"yonseivnl/vlm-rlaif","url":"https://github.com/yonseivnl/vlm-rlaif"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26642,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"TS-LLaVA-34B","metrics":{"mean":"3.38"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26643,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"PLLaVA-34B","metrics":{"Consistency":"3.25","Contextual Understanding":"3.90","Correctness of Information":"3.60","Detail Orientation":"3.20","Temporal Understanding":"2.67","mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26644,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"PPLLaVA-7B","metrics":{"Consistency":"3.20","Contextual Understanding":"3.88","Correctness of Information":"3.32","Detail Orientation":"3.20","Temporal Understanding":"3.0","mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26645,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"SlowFast-LLaVA-34B","metrics":{"mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26646,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VideoGPT+","metrics":{"Consistency":"3.39","Contextual Understanding":"3.74","Correctness of Information":"3.27","Detail Orientation":"3.18","Temporal Understanding":"2.83","mean":"3.28"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26647,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"IG-VLM-GPT4v","metrics":{"Consistency":"3.13","Contextual Understanding":"3.61","Correctness of Information":"3.40","Detail Orientation":"2.80","Temporal Understanding":"2.89","mean":"3.17"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26648,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"ST-LLM-7B","metrics":{"Consistency":"2.81","Contextual Understanding":"3.74","Correctness of Information":"3.23","Detail Orientation":"3.05","Temporal Understanding":"2.93","mean":"3.15"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26649,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VideoChat2_HD_mistral","metrics":{"Consistency":"2.84","Contextual Understanding":"3.72","Correctness of Information":"3.40","Detail Orientation":"2.91","Temporal Understanding":"2.65","mean":"3.10"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26650,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"CAT-7B","metrics":{"Consistency":"2.89","Contextual Understanding":"3.49","Correctness of Information":"3.08","Detail Orientation":"2.95","Temporal Understanding":"2.81","mean":"3.07"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26651,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"LITA-13B","metrics":{"Consistency":"3.19","Contextual Understanding":"3.43","Correctness of Information":"2.94","Detail Orientation":"2.98","Temporal Understanding":"2.68","mean":"3.04"},"paper_url":"https://arxiv.org/abs/2403.19046v1","paper_title":"LITA: Language Instructed Temporal-Localization Assistant","paper_date":"2024-03-27","code_links":[{"title":"nvlabs/lita","url":"https://github.com/nvlabs/lita"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26652,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"LLaMA-VID-13B (2 Token)","metrics":{"Consistency":"2.63","Contextual Understanding":"3.60","Correctness of Information":"3.07","Detail Orientation":"3.05","Temporal Understanding":"2.58","mean":"2.99"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26653,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"Chat-UniVi","metrics":{"Consistency":"2.81","Contextual Understanding":"3.46","Correctness of Information":"2.89","Detail Orientation":"2.91","Temporal Understanding":"2.39","mean":"2.99"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26654,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VideoChat2","metrics":{"Consistency":"2.81","Contextual Understanding":"3.51","Correctness of Information":"3.02","Detail Orientation":"2.88","Temporal Understanding":"2.66","mean":"2.98"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26655,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"LLaMA-VID-7B (2 Token)","metrics":{"Consistency":"2.51","Contextual Understanding":"3.53","Correctness of Information":"2.96","Detail Orientation":"3.00","Temporal Understanding":"2.46","mean":"2.89"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26656,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VTimeLLM","metrics":{"Consistency":"2.47","Contextual Understanding":"3.40","Correctness of Information":"2.78","Detail Orientation":"3.10","Temporal Understanding":"2.49","mean":"2.85"},"paper_url":"https://arxiv.org/abs/2311.18445v1","paper_title":"VTimeLLM: Empower LLM to Grasp Video Moments","paper_date":"2023-11-30","code_links":[{"title":"huangb23/vtimellm","url":"https://github.com/huangb23/vtimellm"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26657,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"BT-Adapter","metrics":{"Consistency":"2.46","Contextual Understanding":"3.27","Correctness of Information":"2.68","Detail Orientation":"2.69","Temporal Understanding":"2.34","mean":"2.69"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26658,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"BT-Adapter (zero-shot)","metrics":{"Consistency":"2.2","Contextual Understanding":"2.89","Correctness of Information":"2.16","Detail Orientation":"2.46","Temporal Understanding":"2.13","mean":"2.46"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26659,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"Video-ChatGPT","metrics":{"Consistency":"2.37","Contextual Understanding":"2.62","Correctness of Information":"2.4","Detail Orientation":"2.52","Temporal Understanding":"1.98","mean":"2.38"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26660,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"Video Chat","metrics":{"Consistency":"2.24","Contextual Understanding":"2.53","Correctness of Information":"2.23","Detail Orientation":"2.50","Temporal Understanding":"1.94","mean":"2.29"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26661,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"LLaMA Adapter","metrics":{"Consistency":"2.15","Contextual Understanding":"2.30","Correctness of Information":"2.03","Detail Orientation":"2.32","Temporal Understanding":"1.98","mean":"2.16"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":26662,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"Video LLaMA","metrics":{"Consistency":"1.79","Contextual Understanding":"2.16","Correctness of Information":"1.96","Detail Orientation":"2.18","Temporal Understanding":"1.82","mean":"1.98"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134324,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"PPLLaVA-7B-dpo","metrics":{"Consistency":"3.81","Contextual Understanding":"4.21","Correctness of Information":"3.85","Detail Orientation":"3.56","Temporal Understanding":"3.21","mean":"3.73"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134325,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VLM-RLAIF","metrics":{"Consistency":"3.32","Contextual Understanding":"4","Correctness of Information":"3.63","Detail Orientation":"3.25","Temporal Understanding":"3.23","mean":"3.49"},"paper_url":"https://arxiv.org/abs/2402.03746v3","paper_title":"Tuning Large Multimodal Models for Videos using Reinforcement Learning from AI Feedback","paper_date":"2024-02-06","code_links":[{"title":"yonseivnl/vlm-rlaif","url":"https://github.com/yonseivnl/vlm-rlaif"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134326,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"TS-LLaVA-34B","metrics":{"mean":"3.38"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134327,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"PLLaVA-34B","metrics":{"Consistency":"3.25","Contextual Understanding":"3.90","Correctness of Information":"3.60","Detail Orientation":"3.20","Temporal Understanding":"2.67","mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134328,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"PPLLaVA-7B","metrics":{"Consistency":"3.20","Contextual Understanding":"3.88","Correctness of Information":"3.32","Detail Orientation":"3.20","Temporal Understanding":"3.0","mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134329,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"SlowFast-LLaVA-34B","metrics":{"mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134330,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VideoGPT+","metrics":{"Consistency":"3.39","Contextual Understanding":"3.74","Correctness of Information":"3.27","Detail Orientation":"3.18","Temporal Understanding":"2.83","mean":"3.28"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134331,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"IG-VLM-GPT4v","metrics":{"Consistency":"3.13","Contextual Understanding":"3.61","Correctness of Information":"3.40","Detail Orientation":"2.80","Temporal Understanding":"2.89","mean":"3.17"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134332,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"ST-LLM-7B","metrics":{"Consistency":"2.81","Contextual Understanding":"3.74","Correctness of Information":"3.23","Detail Orientation":"3.05","Temporal Understanding":"2.93","mean":"3.15"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134333,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VideoChat2_HD_mistral","metrics":{"Consistency":"2.84","Contextual Understanding":"3.72","Correctness of Information":"3.40","Detail Orientation":"2.91","Temporal Understanding":"2.65","mean":"3.10"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134334,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"CAT-7B","metrics":{"Consistency":"2.89","Contextual Understanding":"3.49","Correctness of Information":"3.08","Detail Orientation":"2.95","Temporal Understanding":"2.81","mean":"3.07"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134335,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"LITA-13B","metrics":{"Consistency":"3.19","Contextual Understanding":"3.43","Correctness of Information":"2.94","Detail Orientation":"2.98","Temporal Understanding":"2.68","mean":"3.04"},"paper_url":"https://arxiv.org/abs/2403.19046v1","paper_title":"LITA: Language Instructed Temporal-Localization Assistant","paper_date":"2024-03-27","code_links":[{"title":"nvlabs/lita","url":"https://github.com/nvlabs/lita"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134336,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"LLaMA-VID-13B (2 Token)","metrics":{"Consistency":"2.63","Contextual Understanding":"3.60","Correctness of Information":"3.07","Detail Orientation":"3.05","Temporal Understanding":"2.58","mean":"2.99"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134337,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"Chat-UniVi","metrics":{"Consistency":"2.81","Contextual Understanding":"3.46","Correctness of Information":"2.89","Detail Orientation":"2.91","Temporal Understanding":"2.39","mean":"2.99"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134338,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VideoChat2","metrics":{"Consistency":"2.81","Contextual Understanding":"3.51","Correctness of Information":"3.02","Detail Orientation":"2.88","Temporal Understanding":"2.66","mean":"2.98"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134339,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"LLaMA-VID-7B (2 Token)","metrics":{"Consistency":"2.51","Contextual Understanding":"3.53","Correctness of Information":"2.96","Detail Orientation":"3.00","Temporal Understanding":"2.46","mean":"2.89"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134340,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"VTimeLLM","metrics":{"Consistency":"2.47","Contextual Understanding":"3.40","Correctness of Information":"2.78","Detail Orientation":"3.10","Temporal Understanding":"2.49","mean":"2.85"},"paper_url":"https://arxiv.org/abs/2311.18445v1","paper_title":"VTimeLLM: Empower LLM to Grasp Video Moments","paper_date":"2023-11-30","code_links":[{"title":"huangb23/vtimellm","url":"https://github.com/huangb23/vtimellm"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134341,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"BT-Adapter","metrics":{"Consistency":"2.46","Contextual Understanding":"3.27","Correctness of Information":"2.68","Detail Orientation":"2.69","Temporal Understanding":"2.34","mean":"2.69"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134342,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"BT-Adapter (zero-shot)","metrics":{"Consistency":"2.2","Contextual Understanding":"2.89","Correctness of Information":"2.16","Detail Orientation":"2.46","Temporal Understanding":"2.13","mean":"2.46"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134343,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"Video-ChatGPT","metrics":{"Consistency":"2.37","Contextual Understanding":"2.62","Correctness of Information":"2.4","Detail Orientation":"2.52","Temporal Understanding":"1.98","mean":"2.38"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134344,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"Video Chat","metrics":{"Consistency":"2.24","Contextual Understanding":"2.53","Correctness of Information":"2.23","Detail Orientation":"2.50","Temporal Understanding":"1.94","mean":"2.29"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134345,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"LLaMA Adapter","metrics":{"Consistency":"2.15","Contextual Understanding":"2.30","Correctness of Information":"2.03","Detail Orientation":"2.32","Temporal Understanding":"1.98","mean":"2.16"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134346,"task":"Video-based Generative Performance Benchmarking","parent_task":"Generative Visual Question Answering","dataset":"VideoInstruct","model_name":"Video LLaMA","metrics":{"Consistency":"1.79","Contextual Understanding":"2.16","Correctness of Information":"1.96","Detail Orientation":"2.18","Temporal Understanding":"1.82","mean":"1.98"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134455,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"PPLLaVA-7B-dpo","metrics":{"Consistency":"3.81","Contextual Understanding":"4.21","Correctness of Information":"3.85","Detail Orientation":"3.56","Temporal Understanding":"3.21","mean":"3.73"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134456,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"VLM-RLAIF","metrics":{"Consistency":"3.32","Contextual Understanding":"4","Correctness of Information":"3.63","Detail Orientation":"3.25","Temporal Understanding":"3.23","mean":"3.49"},"paper_url":"https://arxiv.org/abs/2402.03746v3","paper_title":"Tuning Large Multimodal Models for Videos using Reinforcement Learning from AI Feedback","paper_date":"2024-02-06","code_links":[{"title":"yonseivnl/vlm-rlaif","url":"https://github.com/yonseivnl/vlm-rlaif"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134457,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"TS-LLaVA-34B","metrics":{"mean":"3.38"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134458,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"PLLaVA-34B","metrics":{"Consistency":"3.25","Contextual Understanding":"3.90","Correctness of Information":"3.60","Detail Orientation":"3.20","Temporal Understanding":"2.67","mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134459,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"PPLLaVA-7B","metrics":{"Consistency":"3.20","Contextual Understanding":"3.88","Correctness of Information":"3.32","Detail Orientation":"3.20","Temporal Understanding":"3.0","mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134460,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"SlowFast-LLaVA-34B","metrics":{"mean":"3.32"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134461,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"VideoGPT+","metrics":{"Consistency":"3.39","Contextual Understanding":"3.74","Correctness of Information":"3.27","Detail Orientation":"3.18","Temporal Understanding":"2.83","mean":"3.28"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134462,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"IG-VLM-GPT4v","metrics":{"Consistency":"3.13","Contextual Understanding":"3.61","Correctness of Information":"3.40","Detail Orientation":"2.80","Temporal Understanding":"2.89","mean":"3.17"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134463,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"ST-LLM-7B","metrics":{"Consistency":"2.81","Contextual Understanding":"3.74","Correctness of Information":"3.23","Detail Orientation":"3.05","Temporal Understanding":"2.93","mean":"3.15"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134464,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"VideoChat2_HD_mistral","metrics":{"Consistency":"2.84","Contextual Understanding":"3.72","Correctness of Information":"3.40","Detail Orientation":"2.91","Temporal Understanding":"2.65","mean":"3.10"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134465,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"CAT-7B","metrics":{"Consistency":"2.89","Contextual Understanding":"3.49","Correctness of Information":"3.08","Detail Orientation":"2.95","Temporal Understanding":"2.81","mean":"3.07"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134466,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"LITA-13B","metrics":{"Consistency":"3.19","Contextual Understanding":"3.43","Correctness of Information":"2.94","Detail Orientation":"2.98","Temporal Understanding":"2.68","mean":"3.04"},"paper_url":"https://arxiv.org/abs/2403.19046v1","paper_title":"LITA: Language Instructed Temporal-Localization Assistant","paper_date":"2024-03-27","code_links":[{"title":"nvlabs/lita","url":"https://github.com/nvlabs/lita"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134467,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"LLaMA-VID-13B (2 Token)","metrics":{"Consistency":"2.63","Contextual Understanding":"3.60","Correctness of Information":"3.07","Detail Orientation":"3.05","Temporal Understanding":"2.58","mean":"2.99"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134468,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"Chat-UniVi","metrics":{"Consistency":"2.81","Contextual Understanding":"3.46","Correctness of Information":"2.89","Detail Orientation":"2.91","Temporal Understanding":"2.39","mean":"2.99"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134469,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"VideoChat2","metrics":{"Consistency":"2.81","Contextual Understanding":"3.51","Correctness of Information":"3.02","Detail Orientation":"2.88","Temporal Understanding":"2.66","mean":"2.98"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134470,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"LLaMA-VID-7B (2 Token)","metrics":{"Consistency":"2.51","Contextual Understanding":"3.53","Correctness of Information":"2.96","Detail Orientation":"3.00","Temporal Understanding":"2.46","mean":"2.89"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134471,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"VTimeLLM","metrics":{"Consistency":"2.47","Contextual Understanding":"3.40","Correctness of Information":"2.78","Detail Orientation":"3.10","Temporal Understanding":"2.49","mean":"2.85"},"paper_url":"https://arxiv.org/abs/2311.18445v1","paper_title":"VTimeLLM: Empower LLM to Grasp Video Moments","paper_date":"2023-11-30","code_links":[{"title":"huangb23/vtimellm","url":"https://github.com/huangb23/vtimellm"}],"metrics_order":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134472,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"BT-Adapter","metrics":{"Consistency":"2.46","Contextual Understanding":"3.27","Correctness of Information":"2.68","Detail Orientation":"2.69","Temporal Understanding":"2.34","mean":"2.69"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134473,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"BT-Adapter (zero-shot)","metrics":{"Consistency":"2.2","Contextual Understanding":"2.89","Correctness of Information":"2.16","Detail Orientation":"2.46","Temporal Understanding":"2.13","mean":"2.46"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134474,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"Video-ChatGPT","metrics":{"Consistency":"2.37","Contextual Understanding":"2.62","Correctness of Information":"2.4","Detail Orientation":"2.52","Temporal Understanding":"1.98","mean":"2.38"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134475,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"Video Chat","metrics":{"Consistency":"2.24","Contextual Understanding":"2.53","Correctness of Information":"2.23","Detail Orientation":"2.50","Temporal Understanding":"1.94","mean":"2.29"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134476,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"LLaMA Adapter","metrics":{"Consistency":"2.15","Contextual Understanding":"2.30","Correctness of Information":"2.03","Detail Orientation":"2.32","Temporal Understanding":"1.98","mean":"2.16"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]},{"id":134477,"task":"Video-based Generative Performance Benchmarking","parent_task":null,"dataset":"VideoInstruct","model_name":"Video LLaMA","metrics":{"Consistency":"1.79","Contextual Understanding":"2.16","Correctness of Information":"1.96","Detail Orientation":"2.18","Temporal Understanding":"1.82","mean":"1.98"},"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":"[\"mean\", \"Correctness of Information\", \"Detail Orientation\", \"Contextual Understanding\", \"Temporal Understanding\", \"Consistency\"]","area":"Reasoning","uses_additional_data":0,"source":"archive","tags":[]}]}