{"url":"/sota/video-based-generative-performance-2","task":{"name":"Video-based Generative Performance Benchmarking (Consistency)","url":"/task/video-based-generative-performance-5","note":null},"dataset":{"name":"VideoInstruct","url":"/dataset/videoinstruct"},"category":"Reasoning","categories":["Reasoning"],"category_note":null,"description":"The benchmark evaluates a generative Video Conversational Model with respect to Consistency.\r\n\r\nWe curate a test set based on the ActivityNet-200 dataset, featuring videos with rich, dense descriptive captions and associated question-answer pairs from human annotations. We develop an evaluation pipeline using the GPT-3.5 model that assigns a relative score to the generated predictions on a scale of 1-5.","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["gpt-score"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"gpt-score":"higher"}},"counts":{"rows":18,"rows_with_code":18,"rows_with_paper_page":18,"rows_dated":18,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"PPLLaVA-7B","metrics":{"gpt-score":"3.81"},"uses_additional_data":false,"paper_date":"2024-11-04","paper":"/paper/ppllava-varied-video-sequence-understanding","paper_url":"https://arxiv.org/abs/2411.02327v2","paper_title":"PPLLaVA: Varied Video Sequence Understanding With Prompt Guidance","code":"https://github.com/farewellthree/ppllava","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":7,"n_samples":9,"n_pointer_only_licence":3}},{"rank_in_archive_order":2,"model":"TS-LLaVA-34B","metrics":{"gpt-score":"3.69"},"uses_additional_data":false,"paper_date":"2024-11-17","paper":"/paper/ts-llava-constructing-visual-tokens-through","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","code":"https://github.com/tingyu215/ts-llava","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":6,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"SlowFast-LLaVA-34B","metrics":{"gpt-score":"3.57"},"uses_additional_data":false,"paper_date":"2024-07-22","paper":"/paper/slowfast-llava-a-strong-training-free","paper_url":"https://arxiv.org/abs/2407.15841v2","paper_title":"SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models","code":"https://github.com/apple/ml-slowfast-llava","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":6}},{"rank_in_archive_order":4,"model":"VideoGPT+","metrics":{"gpt-score":"3.39"},"uses_additional_data":false,"paper_date":"2024-06-13","paper":"/paper/videogpt-integrating-image-and-video-encoders","paper_url":"https://arxiv.org/abs/2406.09418v1","paper_title":"VideoGPT+: Integrating Image and Video Encoders for Enhanced Video Understanding","code":"https://github.com/mbzuai-oryx/videogpt-plus","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":2,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":5,"model":"PLLaVA-34B","metrics":{"gpt-score":"3.25"},"uses_additional_data":false,"paper_date":"2024-04-25","paper":"/paper/pllava-parameter-free-llava-extension-from-1","paper_url":"https://arxiv.org/abs/2404.16994v2","paper_title":"PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning","code":"https://github.com/magic-research/PLLaVA","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":6,"model":"Chat-UniVi","metrics":{"gpt-score":"2.81"},"uses_additional_data":false,"paper_date":"2023-11-14","paper":"/paper/chat-univi-unified-visual-representation","paper_url":"https://arxiv.org/abs/2311.08046v3","paper_title":"Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding","code":"https://github.com/pku-yuangroup/chat-univi","n_code_links":4,"syntology":null},{"rank_in_archive_order":7,"model":"VideoChat2","metrics":{"gpt-score":"2.81"},"uses_additional_data":false,"paper_date":"2023-11-28","paper":"/paper/mvbench-a-comprehensive-multi-modal-video","paper_url":"https://arxiv.org/abs/2311.17005v4","paper_title":"MVBench: A Comprehensive Multi-modal Video Understanding Benchmark","code":"https://github.com/opengvlab/ask-anything","n_code_links":3,"syntology":{"n_ran":7,"n_unverified":3,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"ST-LLM","metrics":{"gpt-score":"2.81"},"uses_additional_data":false,"paper_date":"2024-03-30","paper":"/paper/st-llm-large-language-models-are-effective-1","paper_url":"https://arxiv.org/abs/2404.00308v1","paper_title":"ST-LLM: Large Language Models Are Effective Temporal Learners","code":"https://github.com/TencentARC/ST-LLM","n_code_links":1,"syntology":{"n_ran":7,"n_unverified":4,"n_samples":11,"n_pointer_only_licence":3}},{"rank_in_archive_order":9,"model":"MiniGPT4-video-7B","metrics":{"gpt-score":"2.67"},"uses_additional_data":false,"paper_date":"2024-04-04","paper":"/paper/minigpt4-video-advancing-multimodal-llms-for","paper_url":"https://arxiv.org/abs/2404.03413v1","paper_title":"MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens","code":"https://github.com/Vision-CAIR/MiniGPT4-video","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"VideoChat2_HD_mistral","metrics":{"gpt-score":"2.62"},"uses_additional_data":false,"paper_date":"2023-11-28","paper":"/paper/mvbench-a-comprehensive-multi-modal-video","paper_url":"https://arxiv.org/abs/2311.17005v4","paper_title":"MVBench: A Comprehensive Multi-modal Video Understanding Benchmark","code":"https://github.com/opengvlab/ask-anything","n_code_links":3,"syntology":{"n_ran":7,"n_unverified":3,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"VTimeLLM","metrics":{"gpt-score":"2.47"},"uses_additional_data":false,"paper_date":"2023-11-30","paper":"/paper/vtimellm-empower-llm-to-grasp-video-moments","paper_url":"https://arxiv.org/abs/2311.18445v1","paper_title":"VTimeLLM: Empower LLM to Grasp Video Moments","code":"https://github.com/huangb23/vtimellm","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":6,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":12,"model":"BT-Adapter","metrics":{"gpt-score":"2.46"},"uses_additional_data":false,"paper_date":"2023-09-27","paper":"/paper/one-for-all-video-conversation-is-feasible","paper_url":"https://arxiv.org/abs/2309.15785v2","paper_title":"BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning","code":"https://github.com/farewellthree/BT-Adapter","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":3,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"MovieChat","metrics":{"gpt-score":"2.42"},"uses_additional_data":false,"paper_date":"2023-07-31","paper":"/paper/moviechat-from-dense-token-to-sparse-memory","paper_url":"https://arxiv.org/abs/2307.16449v4","paper_title":"MovieChat: From Dense Token to Sparse Memory for Long Video Understanding","code":"https://github.com/rese1f/MovieChat","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"Video-ChatGPT","metrics":{"gpt-score":"2.37"},"uses_additional_data":false,"paper_date":"2023-06-08","paper":"/paper/video-chatgpt-towards-detailed-video","paper_url":"https://arxiv.org/abs/2306.05424v2","paper_title":"Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models","code":"https://github.com/mbzuai-oryx/video-chatgpt","n_code_links":2,"syntology":null},{"rank_in_archive_order":15,"model":"Video Chat","metrics":{"gpt-score":"2.24"},"uses_additional_data":false,"paper_date":"2023-05-10","paper":"/paper/videochat-chat-centric-video-understanding","paper_url":"https://arxiv.org/abs/2305.06355v2","paper_title":"VideoChat: Chat-Centric Video Understanding","code":"https://github.com/opengvlab/ask-anything","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"BT-Adapter (zero-shot)","metrics":{"gpt-score":"2.2"},"uses_additional_data":false,"paper_date":"2023-09-27","paper":"/paper/one-for-all-video-conversation-is-feasible","paper_url":"https://arxiv.org/abs/2309.15785v2","paper_title":"BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning","code":"https://github.com/farewellthree/BT-Adapter","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":3,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"LLaMA Adapter","metrics":{"gpt-score":"2.15"},"uses_additional_data":false,"paper_date":"2023-04-28","paper":"/paper/llama-adapter-v2-parameter-efficient-visual","paper_url":"https://arxiv.org/abs/2304.15010v1","paper_title":"LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model","code":"https://github.com/opengvlab/llama-adapter","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":18,"model":"Video LLaMA","metrics":{"gpt-score":"1.79"},"uses_additional_data":false,"paper_date":"2023-06-05","paper":"/paper/video-llama-an-instruction-tuned-audio-visual","paper_url":"https://arxiv.org/abs/2306.02858v4","paper_title":"Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding","code":"https://github.com/damo-nlp-sg/video-llama","n_code_links":4,"syntology":{"n_ran":18,"n_unverified":7,"n_samples":25,"n_pointer_only_licence":9}}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":14,"rows_with_any_sample_ran":12,"distinct_papers_with_graph_line":12,"distinct_papers_with_any_sample_ran":10,"samples_over_distinct_papers":{"n_ran":57,"n_unverified":43,"n_samples":100,"n_pointer_only_licence":43,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":65,"n_unverified":49,"n_samples":114,"n_pointer_only_licence":43,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}