{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T09:59:56Z","timestamp":1777888796488,"version":"3.51.4"},"reference-count":86,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.02192","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"23614-23625","source":"Crossref","is-referenced-by-count":1,"title":["FOLDER: Accelerating Multi-Modal Large Language Models with Enhanced Performance"],"prefix":"10.1109","author":[{"given":"Haicheng","family":"Wang","sequence":"first","affiliation":[{"name":"Shanghai Jiao Tong University,Paris Elite Institute of Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhemeng","family":"Yu","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University,Paris Elite Institute of Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gabriele","family":"Spadaro","sequence":"additional","affiliation":[{"name":"Institut Polytechnique de Paris,LTCI, Télécom Paris"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chen","family":"Ju","sequence":"additional","affiliation":[{"name":"Taobao & Tmall Group of Alibaba"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Victor","family":"Qu\u00e9tu","sequence":"additional","affiliation":[{"name":"Institut Polytechnique de Paris,LTCI, Télécom Paris"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shuai","family":"Xiao","sequence":"additional","affiliation":[{"name":"Taobao & Tmall Group of Alibaba"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Enzo","family":"Tartaglione","sequence":"additional","affiliation":[{"name":"Institut Polytechnique de Paris,LTCI, Télécom Paris"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv preprint"},{"key":"ref2","article-title":"Anthropic","year":"2024","journal-title":"The claude 3 model family: Opus, sonnet, haiku"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73397-0_15"},{"key":"ref4","article-title":"Openflamingo: An opensource framework for training large autoregressive visionlanguage models","author":"Awadalla","year":"2023","journal-title":"arXiv preprint"},{"key":"ref5","article-title":"Butterfly effects of sgd noise: Error amplification in behavior cloning and autoregression","author":"Block","year":"2023","journal-title":"arXiv preprint"},{"key":"ref6","article-title":"Token merging: Your vit but faster","author":"Bolya","year":"2022","journal-title":"arXiv preprint"},{"key":"ref7","article-title":"Minigpt-v2: large language model as a unified interface for vision-language multi-task learning","author":"Chen","year":"2023","journal-title":"arXiv preprint"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0850"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73004-7_2"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4231-5"},{"key":"ref11","first-page":"24185","article-title":"Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Chen"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3581783.3611987"},{"key":"ref13","author":"Dao","year":"2023","journal-title":"Flashattention-2: Faster attention with better parallelism and work partitioning"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1189"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref16","article-title":"An image is worth $16 \\times 16$ words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020","journal-title":"arXiv preprint"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3664647.3685520"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/BF02288367"},{"key":"ref19","article-title":"Mmbench-video: A long-form multi-shot benchmark for holistic video understanding","author":"Fang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00146"},{"key":"ref21","article-title":"ptq: Accurate post-training quantization for generative pre-trained transformers","author":"Frantar","year":"2022","journal-title":"arXiv preprint"},{"key":"ref22","article-title":"Mme: A comprehensive evaluation benchmark for multimodal large language models","author":"Fu","year":"2023","journal-title":"arXiv preprint"},{"key":"ref23","article-title":"Video-mme: The first-ever comprehensive evaluation benchmark of multi-modal 11 ms in video analysis","author":"Fu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref24","article-title":"Vita-1.5: Towards gpt-4o level real-time vision and speech interaction","author":"Fu","year":"2025","journal-title":"arXiv preprint"},{"key":"ref25","article-title":"Grok","volume-title":"Grok1.5 homepage","year":"2024"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01363"},{"key":"ref27","article-title":"Rethinking token reduction in mllms: Towards a unified paradigm for training-free acceleration","author":"Han","year":"2024","journal-title":"arXiv preprint"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"ref29","article-title":"Dynamic-llava: Efficient multimodal large language models via dynamic vision-language context sparsification","author":"Huang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01300"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_25"},{"key":"ref32","article-title":"Brave: Broadening the visual encoding of vision-language models","author":"Kar","year":"2024","journal-title":"arXiv preprint"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01441"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20083-0_37"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2789"},{"key":"ref36","article-title":"Seed-bench: Benchmarking multimodal 11 ms with generative comprehension","author":"Li","year":"2023","journal-title":"arXiv preprint"},{"key":"ref37","article-title":"Llava-onevision: Easy visual task transfer","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref38","first-page":"12888","article-title":"Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation","volume-title":"International conference on machine learning","author":"Li","year":"2022"},{"key":"ref39","first-page":"19730","article-title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"International conference on machine learning","author":"Li","year":"2023"},{"key":"ref40","article-title":"Videochat: Chat-centric video understanding","author":"Li","year":"2023","journal-title":"arXiv preprint"},{"key":"ref41","author":"Harold Li","year":"2019","journal-title":"Visualbert: A simple and performant baseline for vision and language"},{"key":"ref42","article-title":"Tokenpacker: Efficient visual projector for multimodal llm","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02240"},{"key":"ref44","author":"Liang","year":"2022","journal-title":"Not all patches are what you need: Expediting vision transformers via token reorganizations"},{"key":"ref45","article-title":"Video-llava: Learning united visual representation by alignment before projection","author":"Lin","year":"2023","journal-title":"arXiv preprint"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02520"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref49","author":"Liu","year":"2023","journal-title":"Ring attention with blockwise transformers for near-infinite context"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref51","article-title":"Multi-stage vision token dropping: Towards efficient multimodal large language model","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref52","article-title":"Tempcompass: Do video $l l ~\\text{ms}$ really understand videos?","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref53","article-title":"Points1. 5: Building a visionlanguage model towards real world applications","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.00390"},{"key":"ref55","article-title":"Learn to explain: Multimodal reasoning via thought chains for science question answering","volume-title":"The 36th Conference on Neural Information Processing Systems (NeurIPS)","author":"Lu","year":"2022"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/WACV51458.2022.00264"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/ICDAR.2019.00156"},{"key":"ref58","article-title":"Dinov2: Learning robust visual features without supervision","author":"Oquab","year":"2023","journal-title":"arXiv preprint"},{"key":"ref59","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"International conference on machine learning","author":"Radford","year":"2021"},{"key":"ref60","article-title":"Dynamicvit: Efficient vision transformers with dynamic token sparsification","author":"Rao","year":"2021","journal-title":"NeurIPS"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1023\/A:1026543900054"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20074-8_9"},{"key":"ref63","author":"Shang","year":"2024","journal-title":"Llava-prumerge: Adaptive token reduction for efficient large multimodal models. arXiv preprint"},{"key":"ref64","article-title":"Upop: Unified and progressive pruning for compressing vision-language transformers","author":"Shi","year":"2023","journal-title":"arXiv preprint"},{"key":"ref65","article-title":"Crossget: Cross-guided ensemble of tokens for accelerating vision-language transformers","author":"Shi","year":"2023","journal-title":"arXiv preprint"},{"key":"ref66","article-title":"Eagle: Exploring the design space for multimodal 11 ms with mixture of encoders","author":"Shi","year":"2024","journal-title":"arXiv preprint"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01519"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1514"},{"key":"ref69","article-title":"Gemini: a family of highly capable multimodal models","author":"Team","year":"2023","journal-title":"arXiv preprint"},{"key":"ref70","first-page":"9568","article-title":"Eyes wide shut? exploring the visual shortcomings of multimodal 11 ms","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Tong","year":"2024"},{"key":"ref71","article-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution","author":"Wang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref72","article-title":"Efficientvlm: Fast and accurate vision-language models via knowledge distillation and modal-adaptive pruning","author":"Wang","year":"2022","journal-title":"arXiv preprint"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0907"},{"key":"ref74","first-page":"38087","article-title":"Smoothquant: Accurate and efficient post-training quantization for large language models","volume-title":"International Conference on Machine Learning","author":"Xiao","year":"2023"},{"key":"ref75","article-title":"Pyramiddrop: Accelerating your large vision-language models via pyramid visual redundancy reduction","author":"Xing","year":"2024","journal-title":"arXiv preprint"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72698-9_8"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.01843"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02325"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02777"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref81","author":"Yuan","year":"2023","journal-title":"Mmbench: Is your multi-modal model an all-around player?"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00913"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-demo.49"},{"key":"ref84","article-title":"Sparsevlm: Visual token sparsification for efficient vision-language model inference","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref85","article-title":"Aim: Adaptive inference of multi-modal $l l ~\\text{ms}$ via token merging and pruning","author":"Zhong","year":"2024","journal-title":"arXiv preprint"},{"key":"ref86","article-title":"Mlvu: A comprehensive benchmark for multitask long video understanding","author":"Zhou","year":"2024","journal-title":"arXiv e-prints"}],"event":{"name":"2025 IEEE\/CVF International Conference on Computer Vision (ICCV)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,10,19]]},"end":{"date-parts":[[2025,10,25]]}},"container-title":["2025 IEEE\/CVF International Conference on Computer Vision (ICCV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11443115\/11443287\/11446052.pdf?arnumber=11446052","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:58:27Z","timestamp":1777611507000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11446052\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":86,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.02192","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}