{"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/fastvid-dynamic-density-pruning-for-fast","title":"FastVID: Dynamic Density Pruning for Fast Video Large Language Models","arxiv_id":"2503.11187","date":"2025-03-14","proceeding":null,"authors":["Leqi Shen","Guoqiang Gong","Tao He","Yifeng Zhang","Pengzhang Liu","Sicheng Zhao","Guiguang Ding"],"abstract":"Video Large Language Models have shown impressive capabilities in video comprehension, yet their practical deployment is hindered by substantial inference costs caused by redundant video tokens. Existing pruning techniques fail to fully exploit the spatiotemporal redundancy inherent in video data. To bridge this gap, we perform a systematic analysis of video redundancy from two perspectives: temporal context and visual context. Leveraging this insight, we propose Dynamic Density Pruning for Fast Video LLMs termed FastVID. Specifically, FastVID dynamically partitions videos into temporally ordered segments to preserve temporal structure and applies a density-based token pruning strategy to maintain essential visual information. Our method significantly reduces computational overhead while maintaining temporal and visual integrity. Extensive evaluations show that FastVID achieves state-of-the-art performance across various short- and long-video benchmarks on leading Video LLMs, including LLaVA-OneVision and LLaVA-Video. Notably, FastVID effectively prunes 90% of video tokens while retaining 98.0% of LLaVA-OneVision's original performance. The code is available at https://github.com/LunarShen/FastVID.","url_abs":"https://arxiv.org/abs/2503.11187v1","url_pdf":"https://arxiv.org/pdf/2503.11187v1.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":"fastvid-dynamic-density-pruning-for-fast","repo_url":"https://github.com/lunarshen/fastvid","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"fastvid-dynamic-density-pruning-for-fast","repo_url":"https://github.com/cokeshao/holitom","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"pruning","method_name":"Pruning"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2503.11187","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.11187"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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":"deterministic:regex_extraction","url":"https://github.com/LunarShen/FastVID","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lunarshen/fastvid","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cokeshao/holitom","reach":null}],"summary":{"ran_fixture":3,"ran":1,"ran_draft_wrong":3,"unverified":4},"by_repo_kind":{"official":{"samples":3,"ran":1,"repositories":1},"listed":{"samples":4,"ran":2,"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":4,"samples":[{"code_sha256_prefix":"30d7eec482ebf6b1","entry":"repeat_kv","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":2,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"30d7eec482ebf6b1"}},{"code_sha256_prefix":"53284346fce3ef1c","entry":"MaskedDrop","repo":"lunarshen/fastvid","repo_kind":"official","path":"fastvid_llavaonevision/LLaVA-NeXT/llava/model/multimodal_resampler/masked_drop.py","file_url":"https://github.com/lunarshen/fastvid/blob/HEAD/fastvid_llavaonevision/LLaVA-NeXT/llava/model/multimodal_resampler/masked_drop.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"53284346fce3ef1c"}},{"code_sha256_prefix":"bac65c3dafaec040","entry":"apply_rotary_pos_emb","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"bac65c3dafaec040"}},{"code_sha256_prefix":"dd90994dc116b84b","entry":"get_anyres_image_grid_shape","repo":"cokeshao/holitom","repo_kind":"listed","path":"holitom/llava_arch.py","file_url":"https://github.com/cokeshao/holitom/blob/HEAD/holitom/llava_arch.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"dd90994dc116b84b"}},{"code_sha256_prefix":"b99eea6376d1e212","entry":"rotate_half","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"b99eea6376d1e212"}},{"code_sha256_prefix":"e03d53ba9d4f9ae5","entry":"rotate_half","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"e03d53ba9d4f9ae5"}},{"code_sha256_prefix":"3999ff487573f32c","entry":"select_best_resolution","repo":"cokeshao/holitom","repo_kind":"listed","path":"holitom/llava_arch.py","file_url":"https://github.com/cokeshao/holitom/blob/HEAD/holitom/llava_arch.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3999ff487573f32c"}},{"code_sha256_prefix":"65d341b7890a432b","entry":"LlavaMetaForCausalLM_holitom","repo":"cokeshao/holitom","repo_kind":"listed","path":"holitom/llava_arch.py","file_url":"https://github.com/cokeshao/holitom/blob/HEAD/holitom/llava_arch.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"65d341b7890a432b"}},{"code_sha256_prefix":"24acaeb8ad82253e","entry":"apply_rotary_pos_emb_flashatt","repo":"lunarshen/fastvid","repo_kind":"official","path":"fastvid_qwen25vl/lmms-eval/lmms_eval/models/qwenvlutils/modeling_qwen2_5_vl.py","file_url":"https://github.com/lunarshen/fastvid/blob/HEAD/fastvid_qwen25vl/lmms-eval/lmms_eval/models/qwenvlutils/modeling_qwen2_5_vl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"24acaeb8ad82253e"}},{"code_sha256_prefix":"09445ca10af72017","entry":"apply_rotary_pos_emb_vision","repo":"lunarshen/fastvid","repo_kind":"official","path":"fastvid_qwen25vl/lmms-eval/lmms_eval/models/qwenvlutils/modeling_qwen2_5_vl.py","file_url":"https://github.com/lunarshen/fastvid/blob/HEAD/fastvid_qwen25vl/lmms-eval/lmms_eval/models/qwenvlutils/modeling_qwen2_5_vl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"09445ca10af72017"}},{"code_sha256_prefix":"16c9ecf6f7fea719","entry":"rank0_print","repo":"cokeshao/holitom","repo_kind":"listed","path":"holitom/llava_arch.py","file_url":"https://github.com/cokeshao/holitom/blob/HEAD/holitom/llava_arch.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"16c9ecf6f7fea719"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}