{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/slowfast-llava-a-strong-training-free","title":"SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models","arxiv_id":"2407.15841","date":"2024-07-22","proceeding":null,"authors":["Mingze Xu","Mingfei Gao","Zhe Gan","Hong-You Chen","Zhengfeng Lai","Haiming Gang","Kai Kang","Afshin Dehghan"],"abstract":"We propose SlowFast-LLaVA (or SF-LLaVA for short), a training-free video large language model (LLM) that can jointly capture detailed spatial semantics and long-range temporal context without exceeding the token budget of commonly used LLMs. This is realized by using a two-stream SlowFast design of inputs for Video LLMs to aggregate features from sampled frames in an effective way. Specifically, the Slow pathway extracts features at a low frame rate while keeping as much spatial detail as possible (e.g., with 12x24 tokens), and the Fast pathway operates on a high frame rate but uses a larger spatial pooling stride (e.g., downsampling 6x) to focus on the motion cues. As a result, this design allows us to adequately capture both spatial and temporal features that are beneficial for detailed video understanding. Experimental results show that SF-LLaVA outperforms existing training-free methods on a wide range of video tasks. On some benchmarks, it achieves comparable or even better performance compared to state-of-the-art Video LLMs that are fine-tuned on video datasets. Code has been made available at: https://github.com/apple/ml-slowfast-llava.","url_abs":"https://arxiv.org/abs/2407.15841v2","url_pdf":"https://arxiv.org/pdf/2407.15841v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"slowfast-llava-a-strong-training-free","repo_url":"https://github.com/apple/ml-slowfast-llava","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"large-language-model","task_name":"Large Language Model"},{"task_slug":"video-understanding","task_name":"Video Understanding"},{"task_slug":"video-based-generative-performance","task_name":"Video-based Generative Performance Benchmarking"},{"task_slug":"video-based-generative-performance-5","task_name":"Video-based Generative Performance Benchmarking (Consistency)"},{"task_slug":"video-based-generative-performance-3","task_name":"Video-based Generative Performance Benchmarking (Contextual Understanding)"},{"task_slug":"video-based-generative-performance-1","task_name":"Video-based Generative Performance Benchmarking (Correctness of Information)"},{"task_slug":"video-based-generative-performance-2","task_name":"Video-based Generative Performance Benchmarking (Detail Orientation))"},{"task_slug":"video-based-generative-performance-4","task_name":"Video-based Generative Performance Benchmarking (Temporal Understanding)"},{"task_slug":"zeroshot-video-question-answer","task_name":"Zero-Shot Video Question Answer"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-based-generative-performance","task":"Video-based Generative Performance Benchmarking","dataset":"VideoInstruct","model":"SlowFast-LLaVA-34B","rank_in_archive_order":6,"of":23,"metrics":{"mean":"3.32"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-2","task":"Video-based Generative Performance Benchmarking (Consistency)","dataset":"VideoInstruct","model":"SlowFast-LLaVA-34B","rank_in_archive_order":3,"of":18,"metrics":{"gpt-score":"3.57"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-3","task":"Video-based Generative Performance Benchmarking (Contextual Understanding)","dataset":"VideoInstruct","model":"SlowFast-LLaVA-34B","rank_in_archive_order":4,"of":18,"metrics":{"gpt-score":"3.84"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-1","task":"Video-based Generative Performance Benchmarking (Correctness of Information)","dataset":"VideoInstruct","model":"SlowFast-LLaVA-34B","rank_in_archive_order":4,"of":18,"metrics":{"gpt-score":"3.48"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-4","task":"Video-based Generative Performance Benchmarking (Detail Orientation))","dataset":"VideoInstruct","model":"SlowFast-LLaVA-34B","rank_in_archive_order":8,"of":18,"metrics":{"gpt-score":"2.96"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-5","task":"Video-based Generative Performance Benchmarking (Temporal Understanding)","dataset":"VideoInstruct","model":"SlowFast-LLaVA-34B","rank_in_archive_order":4,"of":18,"metrics":{"gpt-score":"2.77"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-activitynet","task":"Zero-Shot Video Question Answer","dataset":"ActivityNet-QA","model":"SlowFast-LLaVA-34B","rank_in_archive_order":5,"of":28,"metrics":{"Accuracy":"59.2","Confidence Score":"3.5"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-egoschema","task":"Zero-Shot Video Question Answer","dataset":"EgoSchema (subset)","model":"SlowFast-LLaVA-34B","rank_in_archive_order":12,"of":14,"metrics":{"Accuracy":"47.2"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-intentqa","task":"Zero-Shot Video Question Answer","dataset":"IntentQA","model":"SlowFast-LLaVA-34B","rank_in_archive_order":9,"of":13,"metrics":{"Accuracy":"60.1"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-msrvtt-qa","task":"Zero-Shot Video Question Answer","dataset":"MSRVTT-QA","model":"SlowFast-LLaVA-34B","rank_in_archive_order":4,"of":30,"metrics":{"Accuracy":"67.4","Confidence Score":"3.7"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-msvd-qa","task":"Zero-Shot Video Question Answer","dataset":"MSVD-QA","model":"SlowFast-LLaVA-34B","rank_in_archive_order":6,"of":28,"metrics":{"Accuracy":"79.9","Confidence Score":"4.1"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-next-qa","task":"Zero-Shot Video Question Answer","dataset":"NExT-QA","model":"SlowFast-LLaVA-34B","rank_in_archive_order":18,"of":27,"metrics":{"Accuracy":"64.2"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-tgif-qa","task":"Zero-Shot Video Question Answer","dataset":"TGIF-QA","model":"SlowFast-LLaVA-34B","rank_in_archive_order":5,"of":14,"metrics":{"Accuracy":"80.6","Confidence Score":"4.3"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2407.15841","atlas_url":"https://app.syntology.ai/?focus=2407.15841","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.15841"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/apple/ml-slowfast-llava","reach":{"status":"ok","spdx":"NOASSERTION"}}],"summary":{"ran":2,"ran_fixture":2,"unverified":2},"by_repo_kind":{"official":{"samples":6,"ran":4,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":6,"samples":[{"code_sha256_prefix":"7e03b180fa317c9a","entry":"divide_to_patches","repo":"apple/ml-slowfast-llava","repo_kind":"official","path":"slowfast_llava/llava/mm_utils.py","file_url":"https://github.com/apple/ml-slowfast-llava/blob/HEAD/slowfast_llava/llava/mm_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7e03b180fa317c9a"}},{"code_sha256_prefix":"468eedeba67f1b00","entry":"resize_and_pad_image","repo":"apple/ml-slowfast-llava","repo_kind":"official","path":"slowfast_llava/llava/mm_utils.py","file_url":"https://github.com/apple/ml-slowfast-llava/blob/HEAD/slowfast_llava/llava/mm_utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"468eedeba67f1b00"}},{"code_sha256_prefix":"3999ff487573f32c","entry":"select_best_resolution","repo":"apple/ml-slowfast-llava","repo_kind":"official","path":"slowfast_llava/llava/mm_utils.py","file_url":"https://github.com/apple/ml-slowfast-llava/blob/HEAD/slowfast_llava/llava/mm_utils.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3999ff487573f32c"}},{"code_sha256_prefix":"55c32993da87759b","entry":"unpad_image","repo":"apple/ml-slowfast-llava","repo_kind":"official","path":"slowfast_llava/llava/model/llava_arch.py","file_url":"https://github.com/apple/ml-slowfast-llava/blob/HEAD/slowfast_llava/llava/model/llava_arch.py","link_basis":"plan_row","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"55c32993da87759b"}},{"code_sha256_prefix":"37899f22fb191b37","entry":"pretty_print_semaphore","repo":"apple/ml-slowfast-llava","repo_kind":"official","path":"slowfast_llava/llava/utils.py","file_url":"https://github.com/apple/ml-slowfast-llava/blob/HEAD/slowfast_llava/llava/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"37899f22fb191b37"}},{"code_sha256_prefix":"f9939a84b9a65279","entry":"violates_moderation","repo":"apple/ml-slowfast-llava","repo_kind":"official","path":"slowfast_llava/llava/utils.py","file_url":"https://github.com/apple/ml-slowfast-llava/blob/HEAD/slowfast_llava/llava/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f9939a84b9a65279"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}