{"url":"/sota/video-question-answering-on-next-qa","task":{"name":"Video Question Answering","url":"/task/video-question-answering","note":null},"dataset":{"name":"NExT-QA","url":"/dataset/next-qa"},"category":"Computer Vision","categories":["Computer Vision","Reasoning"],"category_note":null,"description":null,"description_from":null,"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":["Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Accuracy":"higher"}},"counts":{"rows":47,"rows_with_code":40,"rows_with_paper_page":47,"rows_dated":47,"rows_using_additional_data":5},"rows":[{"rank_in_archive_order":1,"model":"LinVT-Qwen2-VL\n(7B)","metrics":{"Accuracy":"85.5"},"uses_additional_data":false,"paper_date":"2024-12-06","paper":"/paper/linvt-empower-your-image-level-large-language","paper_url":"https://arxiv.org/abs/2412.05185v2","paper_title":"LinVT: Empower Your Image-level Large Language Model to Understand Videos","code":"https://github.com/gls0425/linvt","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":8,"n_samples":12,"n_pointer_only_licence":12}},{"rank_in_archive_order":2,"model":"InternVL-2.5(8B)","metrics":{"Accuracy":"85.5"},"uses_additional_data":false,"paper_date":"2024-12-06","paper":"/paper/expanding-performance-boundaries-of-open","paper_url":"https://arxiv.org/abs/2412.05271v4","paper_title":"Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling","code":"https://github.com/opengvlab/internvl","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":8,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"VideoLLaMA3(7B)","metrics":{"Accuracy":"84.5"},"uses_additional_data":false,"paper_date":"2025-01-22","paper":"/paper/videollama-3-frontier-multimodal-foundation","paper_url":"https://arxiv.org/abs/2501.13106v3","paper_title":"VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding","code":"https://github.com/damo-nlp-sg/videollama3","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":8,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"PLM-8B","metrics":{"Accuracy":"84.1"},"uses_additional_data":false,"paper_date":"2025-04-17","paper":"/paper/perceptionlm-open-access-data-and-models-for","paper_url":"https://arxiv.org/abs/2504.13180v1","paper_title":"PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding","code":"https://github.com/facebookresearch/perception_models","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"BIMBA-LLaVA-Qwen2-7B","metrics":{"Accuracy":"83.73"},"uses_additional_data":false,"paper_date":"2025-03-12","paper":"/paper/bimba-selective-scan-compression-for-long","paper_url":"https://arxiv.org/abs/2503.09590v1","paper_title":"BIMBA: Selective-Scan Compression for Long-Range Video Question Answering","code":"https://github.com/md-mohaiminul/BIMBA","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"PLM-3B","metrics":{"Accuracy":"83.4"},"uses_additional_data":false,"paper_date":"2025-04-17","paper":"/paper/perceptionlm-open-access-data-and-models-for","paper_url":"https://arxiv.org/abs/2504.13180v1","paper_title":"PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding","code":"https://github.com/facebookresearch/perception_models","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"LLaVA-Video","metrics":{"Accuracy":"83.2"},"uses_additional_data":false,"paper_date":"2024-10-03","paper":"/paper/video-instruction-tuning-with-synthetic-data","paper_url":"https://arxiv.org/abs/2410.02713v2","paper_title":"Video Instruction Tuning With Synthetic Data","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"NVILA(8B)","metrics":{"Accuracy":"82.2"},"uses_additional_data":false,"paper_date":"2024-12-05","paper":"/paper/nvila-efficient-frontier-visual-language","paper_url":"https://arxiv.org/abs/2412.04468v1","paper_title":"NVILA: Efficient Frontier Visual Language Models","code":"https://github.com/nvlabs/vila","n_code_links":2,"syntology":null},{"rank_in_archive_order":9,"model":"Oryx-1.5(7B)","metrics":{"Accuracy":"81.8"},"uses_additional_data":false,"paper_date":"2024-09-19","paper":"/paper/oryx-mllm-on-demand-spatial-temporal","paper_url":"https://arxiv.org/abs/2409.12961v3","paper_title":"Oryx MLLM: On-Demand Spatial-Temporal Understanding at Arbitrary Resolution","code":"https://github.com/oryx-mllm/oryx","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":3,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"Qwen2-VL(7B)","metrics":{"Accuracy":"81.2"},"uses_additional_data":false,"paper_date":"2024-09-18","paper":"/paper/qwen2-vl-enhancing-vision-language-model-s","paper_url":"https://arxiv.org/abs/2409.12191v2","paper_title":"Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution","code":"https://github.com/qwenlm/qwen2-vl","n_code_links":8,"syntology":{"n_ran":12,"n_unverified":0,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"LongVILA(7B)","metrics":{"Accuracy":"80.7"},"uses_additional_data":false,"paper_date":"2024-08-19","paper":"/paper/longvila-scaling-long-context-visual-language","paper_url":"https://arxiv.org/abs/2408.10188v6","paper_title":"LongVILA: Scaling Long-Context Visual Language Models for Long Videos","code":"https://github.com/nvlabs/vila","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"PLM-1B","metrics":{"Accuracy":"80.3"},"uses_additional_data":false,"paper_date":"2025-04-17","paper":"/paper/perceptionlm-open-access-data-and-models-for","paper_url":"https://arxiv.org/abs/2504.13180v1","paper_title":"PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding","code":"https://github.com/facebookresearch/perception_models","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"LLaVA-OV(72B)","metrics":{"Accuracy":"80.2"},"uses_additional_data":false,"paper_date":"2024-08-06","paper":"/paper/llava-onevision-easy-visual-task-transfer","paper_url":"https://arxiv.org/abs/2408.03326v3","paper_title":"LLaVA-OneVision: Easy Visual Task Transfer","code":"https://github.com/evolvinglmms-lab/lmms-eval","n_code_links":2,"syntology":null},{"rank_in_archive_order":14,"model":"VideoChat2_HD_mistral","metrics":{"Accuracy":"79.5"},"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":15,"model":"LLaVA-OV(7B)","metrics":{"Accuracy":"79.4"},"uses_additional_data":false,"paper_date":"2024-08-06","paper":"/paper/llava-onevision-easy-visual-task-transfer","paper_url":"https://arxiv.org/abs/2408.03326v3","paper_title":"LLaVA-OneVision: Easy Visual Task Transfer","code":"https://github.com/evolvinglmms-lab/lmms-eval","n_code_links":2,"syntology":null},{"rank_in_archive_order":16,"model":"LLaVA-NeXT-Interleave(14B)","metrics":{"Accuracy":"79.1"},"uses_additional_data":false,"paper_date":"2024-07-10","paper":"/paper/llava-next-interleave-tackling-multi-image","paper_url":"https://arxiv.org/abs/2407.07895v2","paper_title":"LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Models","code":"https://github.com/LLaVA-VL/LLaVA-NeXT","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":4}},{"rank_in_archive_order":17,"model":"VideoChat2_mistral","metrics":{"Accuracy":"78.6"},"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":18,"model":"mPLUG-Owl3(8B)","metrics":{"Accuracy":"78.6"},"uses_additional_data":false,"paper_date":"2024-08-09","paper":"/paper/mplug-owl3-towards-long-image-sequence","paper_url":"https://arxiv.org/abs/2408.04840v2","paper_title":"mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models","code":"https://github.com/x-plug/mplug-owl","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"LLaVA-NeXT-Interleave(7B)","metrics":{"Accuracy":"78.2"},"uses_additional_data":false,"paper_date":"2024-07-10","paper":"/paper/llava-next-interleave-tackling-multi-image","paper_url":"https://arxiv.org/abs/2407.07895v2","paper_title":"LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Models","code":"https://github.com/LLaVA-VL/LLaVA-NeXT","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":4}},{"rank_in_archive_order":20,"model":"LLaVA-NeXT-Interleave(DPO)","metrics":{"Accuracy":"77.9"},"uses_additional_data":false,"paper_date":"2024-07-10","paper":"/paper/llava-next-interleave-tackling-multi-image","paper_url":"https://arxiv.org/abs/2407.07895v2","paper_title":"LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Models","code":"https://github.com/LLaVA-VL/LLaVA-NeXT","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":4}},{"rank_in_archive_order":21,"model":"Vamos","metrics":{"Accuracy":"77.3"},"uses_additional_data":false,"paper_date":"2023-11-22","paper":"/paper/vamos-versatile-action-models-for-video","paper_url":"https://arxiv.org/abs/2311.13627v3","paper_title":"Vamos: Versatile Action Models for Video Understanding","code":"https://github.com/brown-palm/Vamos","n_code_links":1,"syntology":{"n_ran":7,"n_unverified":1,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"ViLA (3B)","metrics":{"Accuracy":"75.6"},"uses_additional_data":false,"paper_date":"2023-12-13","paper":"/paper/vlap-efficient-video-language-alignment-via","paper_url":"https://arxiv.org/abs/2312.08367v4","paper_title":"ViLA: Efficient Video-Language Alignment for Video Question Answering","code":"https://github.com/xijun-cs/vila","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"VideoLLaMA2.1(7B)","metrics":{"Accuracy":"75.6"},"uses_additional_data":false,"paper_date":"2024-06-11","paper":"/paper/videollama-2-advancing-spatial-temporal","paper_url":"https://arxiv.org/abs/2406.07476v3","paper_title":"VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs","code":"https://github.com/damo-nlp-sg/videollama2","n_code_links":3,"syntology":{"n_ran":9,"n_unverified":8,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":24,"model":"LLaMA-VQA (33B)","metrics":{"Accuracy":"75.5"},"uses_additional_data":false,"paper_date":"2023-10-24","paper":"/paper/large-language-models-are-temporal-and-causal","paper_url":"https://arxiv.org/abs/2310.15747v2","paper_title":"Large Language Models are Temporal and Causal Reasoners for Video Question Answering","code":"https://github.com/mlvlab/Flipped-VQA","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":25,"model":"ViLA (3B, 4 frames)","metrics":{"Accuracy":"74.4"},"uses_additional_data":false,"paper_date":"2023-12-13","paper":"/paper/vlap-efficient-video-language-alignment-via","paper_url":"https://arxiv.org/abs/2312.08367v4","paper_title":"ViLA: Efficient Video-Language Alignment for Video Question Answering","code":"https://github.com/xijun-cs/vila","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"CREMA","metrics":{"Accuracy":"73.9"},"uses_additional_data":false,"paper_date":"2024-02-08","paper":"/paper/crema-multimodal-compositional-video","paper_url":"https://arxiv.org/abs/2402.05889v3","paper_title":"CREMA: Generalizable and Efficient Video-Language Reasoning via Multimodal Modular Fusion","code":"https://github.com/Yui010206/CREMA","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":2,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"SeViLA","metrics":{"Accuracy":"73.8"},"uses_additional_data":false,"paper_date":"2023-05-11","paper":"/paper/self-chained-image-language-model-for-video-1","paper_url":"https://arxiv.org/abs/2305.06988v2","paper_title":"Self-Chained Image-Language Model for Video Localization and Question Answering","code":"https://github.com/yui010206/sevila","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":7}},{"rank_in_archive_order":28,"model":"TCR","metrics":{"Accuracy":"73.5"},"uses_additional_data":false,"paper_date":"2023-12-19","paper":"/paper/text-conditioned-resampler-for-long-form","paper_url":"https://arxiv.org/abs/2312.11897v3","paper_title":"Text-Conditioned Resampler For Long Form Video Understanding","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":29,"model":"LSTP","metrics":{"Accuracy":"72.1"},"uses_additional_data":false,"paper_date":"2024-02-25","paper":"/paper/lstp-language-guided-spatial-temporal-prompt","paper_url":"https://arxiv.org/abs/2402.16050v2","paper_title":"Efficient Temporal Extrapolation of Multimodal Large Language Models with Temporal Grounding Bridge","code":"https://github.com/bigai-nlco/videotgb","n_code_links":2,"syntology":{"n_ran":5,"n_unverified":3,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":30,"model":"Mirasol3B","metrics":{"Accuracy":"72"},"uses_additional_data":false,"paper_date":"2023-11-09","paper":"/paper/mirasol3b-a-multimodal-autoregressive-model","paper_url":"https://arxiv.org/abs/2311.05698v3","paper_title":"Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":31,"model":"VideoChat2","metrics":{"Accuracy":"68.6"},"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":32,"model":"RTQ","metrics":{"Accuracy":"63.2"},"uses_additional_data":false,"paper_date":"2023-12-01","paper":"/paper/rtq-rethinking-video-language-understanding","paper_url":"https://arxiv.org/abs/2312.00347v2","paper_title":"RTQ: Rethinking Video-language Understanding Based on Image-text Model","code":"https://github.com/SCZwangxiao/RTQ-MM2023","n_code_links":2,"syntology":null},{"rank_in_archive_order":33,"model":"HiTeA","metrics":{"Accuracy":"63.1"},"uses_additional_data":true,"paper_date":"2022-12-30","paper":"/paper/hitea-hierarchical-temporal-aware-video","paper_url":"https://arxiv.org/abs/2212.14546v1","paper_title":"HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":34,"model":"CoVGT(PT)","metrics":{"Accuracy":"60.7"},"uses_additional_data":true,"paper_date":"2023-02-27","paper":"/paper/contrastive-video-question-answering-via","paper_url":"https://arxiv.org/abs/2302.13668v2","paper_title":"Contrastive Video Question Answering via Video Graph Transformer","code":"https://github.com/doc-doc/covgt","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":35,"model":"SeViT","metrics":{"Accuracy":"60.6"},"uses_additional_data":false,"paper_date":"2023-01-27","paper":"/paper/semi-parametric-video-grounded-text","paper_url":"https://arxiv.org/abs/2301.11507v1","paper_title":"Semi-Parametric Video-Grounded Text Generation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":36,"model":"ViperGPT(0-shot)","metrics":{"Accuracy":"60.0"},"uses_additional_data":false,"paper_date":"2023-03-14","paper":"/paper/vipergpt-visual-inference-via-python","paper_url":"https://arxiv.org/abs/2303.08128v1","paper_title":"ViperGPT: Visual Inference via Python Execution for Reasoning","code":"https://github.com/cvlab-columbia/viper","n_code_links":1,"syntology":null},{"rank_in_archive_order":37,"model":"CoVGT","metrics":{"Accuracy":"60.0"},"uses_additional_data":false,"paper_date":"2023-02-27","paper":"/paper/contrastive-video-question-answering-via","paper_url":"https://arxiv.org/abs/2302.13668v2","paper_title":"Contrastive Video Question Answering via Video Graph Transformer","code":"https://github.com/doc-doc/covgt","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":38,"model":"GF","metrics":{"Accuracy":"58.83"},"uses_additional_data":false,"paper_date":"2024-01-03","paper":"/paper/glance-and-focus-memory-prompting-for-multi-1","paper_url":"https://arxiv.org/abs/2401.01529v1","paper_title":"Glance and Focus: Memory Prompting for Multi-Event Video Question Answering","code":"https://github.com/byz0e/glance-focus","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":9,"n_samples":18,"n_pointer_only_licence":0}},{"rank_in_archive_order":39,"model":"VFC","metrics":{"Accuracy":"58.6"},"uses_additional_data":true,"paper_date":"2023-04-13","paper":"/paper/verbs-in-action-improving-verb-understanding","paper_url":"https://arxiv.org/abs/2304.06708v1","paper_title":"Verbs in Action: Improving verb understanding in video-language models","code":"https://github.com/google-research/scenic/tree/main/scenic/projects/verbs_in_action","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":40,"model":"ATM","metrics":{"Accuracy":"58.3"},"uses_additional_data":false,"paper_date":"2023-09-05","paper":"/paper/atm-action-temporality-modeling-for-video","paper_url":"https://arxiv.org/abs/2309.02290v1","paper_title":"ATM: Action Temporality Modeling for Video Question Answering","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":41,"model":"MIST","metrics":{"Accuracy":"57.2"},"uses_additional_data":false,"paper_date":"2022-12-19","paper":"/paper/mist-multi-modal-iterative-spatial-temporal","paper_url":"https://arxiv.org/abs/2212.09522v1","paper_title":"MIST: Multi-modal Iterative Spatial-Temporal Transformer for Long-form Video Question Answering","code":"https://github.com/showlab/mist","n_code_links":1,"syntology":null},{"rank_in_archive_order":42,"model":"VGT(PT)","metrics":{"Accuracy":"56.9"},"uses_additional_data":true,"paper_date":"2022-07-12","paper":"/paper/video-graph-transformer-for-video-question","paper_url":"https://arxiv.org/abs/2207.05342v3","paper_title":"Video Graph Transformer for Video Question Answering","code":"https://github.com/sail-sg/vgt","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":5,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":43,"model":"PAXION","metrics":{"Accuracy":"56.9"},"uses_additional_data":true,"paper_date":"2023-05-18","paper":"/paper/paxion-patching-action-knowledge-in-video-1","paper_url":"https://arxiv.org/abs/2305.10683v4","paper_title":"Paxion: Patching Action Knowledge in Video-Language Foundation Models","code":"https://github.com/mikewangwzhl/paxion","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":44,"model":"VGT","metrics":{"Accuracy":"55.0"},"uses_additional_data":false,"paper_date":"2022-07-12","paper":"/paper/video-graph-transformer-for-video-question","paper_url":"https://arxiv.org/abs/2207.05342v3","paper_title":"Video Graph Transformer for Video Question Answering","code":"https://github.com/sail-sg/vgt","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":5,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":45,"model":"ATP","metrics":{"Accuracy":"54.3"},"uses_additional_data":false,"paper_date":"2022-06-03","paper":"/paper/revisiting-the-video-in-video-language","paper_url":"https://arxiv.org/abs/2206.01720v1","paper_title":"Revisiting the \"Video\" in Video-Language Understanding","code":"https://github.com/stanfordvl/atp-video-language","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":2,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":46,"model":"P3D-G","metrics":{"Accuracy":"53.4"},"uses_additional_data":false,"paper_date":"2022-02-18","paper":"/paper/2-5-1-d-spatio-temporal-scene-graphs-for","paper_url":"https://arxiv.org/abs/2202.09277v2","paper_title":"(2.5+1)D Spatio-Temporal Scene Graphs for Video Question Answering","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":47,"model":"HQGA","metrics":{"Accuracy":"51.4"},"uses_additional_data":false,"paper_date":"2021-12-12","paper":"/paper/video-as-conditional-graph-hierarchy-for","paper_url":"https://arxiv.org/abs/2112.06197v2","paper_title":"Video as Conditional Graph Hierarchy for Multi-Granular Question Answering","code":"https://github.com/doc-doc/hqga","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":3,"n_samples":6,"n_pointer_only_licence":0}}],"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 6,821 of the 9,623 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":9623,"papers_checked":6821,"papers_extracted_not_yet_verified":65,"boards_without_verdict":29,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+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":28,"rows_with_any_sample_ran":27,"distinct_papers_with_graph_line":21,"distinct_papers_with_any_sample_ran":20,"samples_over_distinct_papers":{"n_ran":102,"n_unverified":72,"n_samples":174,"n_pointer_only_licence":25,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":134,"n_unverified":87,"n_samples":221,"n_pointer_only_licence":33,"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"}}}