{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,8,21]],"date-time":"2025-08-21T16:24:57Z","timestamp":1755793497694,"version":"3.44.0"},"reference-count":102,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,6,10]],"date-time":"2025-06-10T00:00:00Z","timestamp":1749513600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,6,10]],"date-time":"2025-06-10T00:00:00Z","timestamp":1749513600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001321","name":"National Research Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001321","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,6,10]]},"DOI":"10.1109\/cvpr52734.2025.02708","type":"proceedings-article","created":{"date-parts":[[2025,8,13]],"date-time":"2025-08-13T17:26:42Z","timestamp":1755106002000},"page":"29083-29095","source":"Crossref","is-referenced-by-count":0,"title":["Live: Learning Video LLM with Streaming Speech Transcription at Scale"],"prefix":"10.1109","author":[{"given":"Joya","family":"Chen","sequence":"first","affiliation":[{"name":"National University of Singapore,Show Lab"}]},{"given":"Ziyun","family":"Zeng","sequence":"additional","affiliation":[{"name":"National University of Singapore,Show Lab"}]},{"given":"Yiqi","family":"Lin","sequence":"additional","affiliation":[{"name":"National University of Singapore,Show Lab"}]},{"given":"Wei","family":"Li","sequence":"additional","affiliation":[{"name":"Bytedance"}]},{"given":"Zejun","family":"Ma","sequence":"additional","affiliation":[{"name":"Bytedance"}]},{"given":"Mike Zheng","family":"Shou","sequence":"additional","affiliation":[{"name":"National University of Singapore,Show Lab"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00774"},{"article-title":"Flamingo: a visual language model for few-shot learning","volume-title":"NeurIPS","author":"Alayrac","key":"ref2"},{"article-title":"Gemini: A family of highly capable multimodal models","year":"2023","author":"Anil","key":"ref3"},{"article-title":"Qwen technical report","year":"2023","author":"Bai","key":"ref4"},{"article-title":"Qwen-vl: A frontier large vision-language model with versatile abilities","year":"2023","author":"Bai","key":"ref5"},{"article-title":"Qwen2.5-vl technical report","year":"2024","author":"Bai","key":"ref6"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-78"},{"key":"ref8","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"NeurIPS","author":"Brown"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.675"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.502"},{"article-title":"A short note about kinetics-600","year":"2018","author":"Carreira","key":"ref11"},{"article-title":"Minigpt-v2: large language model as a unified interface for vision-language multi-task learning","year":"2023","author":"Chen","key":"ref12"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01742"},{"article-title":"Sharegpt4video: Improving video understanding and generation with better captions","year":"2024","author":"Chen","key":"ref14"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01265"},{"article-title":"Expanding performance boundaries of open-source multimodal models with model, data, and test-time scaling","year":"2024","author":"Chen","key":"ref16"},{"article-title":"Videollama 2: Advancing spatial-temporal modeling and audio understanding in video-llms","year":"2024","author":"Cheng","key":"ref17"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.747"},{"article-title":"Instructblip: Towards general-purpose vision-language models with instruction tuning","year":"2023","author":"Dai","key":"ref19"},{"article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","volume-title":"ICLR","author":"Dosovitskiy","key":"ref20"},{"article-title":"The llama 3 herd of models","year":"2024","author":"Dubey","key":"ref21"},{"article-title":"Video-mme: The first-ever comprehensive evaluation benchmark of multi-modal llms in video analysis","year":"2024","author":"Fu","key":"ref22"},{"article-title":"Vita: Towards open-source interactive omni multimodal llm","year":"2024","author":"Fu","key":"ref23"},{"article-title":"Vita-1.5: Towards gpt-4o level real-time vision and speech interaction","year":"2025","author":"Fu","key":"ref24"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46454-1_17"},{"year":"2024","key":"ref26","article-title":"Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.622"},{"journal-title":"Hello gpt-4o","year":"2024","key":"ref28"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298698"},{"article-title":"Training compute-optimal large language models","year":"2022","author":"Hoffmann","key":"ref30"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.aacl-main.48"},{"article-title":"Online video understanding: A comprehensive benchmark and memory-augmented method","year":"2025","author":"Huang","key":"ref32"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1016\/j.cviu.2016.10.018"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.01347"},{"article-title":"Scaling laws for neural language models","year":"2020","author":"Kaplan","key":"ref35"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.83"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.00915"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D18-1167"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3571946"},{"article-title":"Llava-onevision: Easy visual task transfer","year":"2024","author":"Li","key":"ref40"},{"article-title":"Llava-onevision: Easy visual task transfer","year":"2024","author":"Li","key":"ref41"},{"article-title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","year":"2023","author":"Li","key":"ref42"},{"article-title":"Videochat: Chat-centric video understanding","year":"2023","author":"Li","key":"ref43"},{"article-title":"Ovo-bench: How far is your video-llms from real-world online video understanding?","year":"2025","author":"Li","key":"ref44"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02196"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.342"},{"article-title":"Streaming-bench: Assessing the gap for mllms to achieve streaming video understanding","year":"2024","author":"Lin","key":"ref47"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02520"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02484"},{"article-title":"Visual instruction tuning","volume-title":"NeurIPS","author":"Liu","key":"ref50"},{"article-title":"Streamchat: Chatting with streaming video","year":"2024","author":"Liu","key":"ref51"},{"article-title":"Oryx mllm: On-demand spatial-temporal understanding at arbitrary resolution","year":"2024","author":"Liu","key":"ref52"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.679"},{"key":"ref54","first-page":"46212","article-title":"Egoschema: A diagnostic benchmark for very long-form video language understanding","volume-title":"NeurIPS","author":"Mangalam"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73397-0_18"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00272"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW59228.2023.00536"},{"volume-title":"Introducing chatgpt","year":"2023","key":"ref58"},{"journal-title":"GPT-4 technical report","year":"2023","key":"ref59"},{"volume-title":"Gpt-4v(ision) system card","year":"2023","key":"ref60"},{"key":"ref61","article-title":"Perception test: A diagnostic benchmark for multimodal video models","volume":"36","author":"Patraucean","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"article-title":"Streaming long video understanding with large language models","volume-title":"NeurIPS","author":"Qian","key":"ref62"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02239"},{"journal-title":"Improving language understanding by generative pre-training","year":"2018","author":"Radford","key":"ref64"},{"journal-title":"Language models are unsupervised multitask learners","year":"2019","author":"Radford","key":"ref65"},{"key":"ref66","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"ICML","author":"Radford"},{"key":"ref67","first-page":"28492","article-title":"Robust speech recognition via large-scale weak supervision","volume-title":"ICML","author":"Radford"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.99"},{"article-title":"Longvu: Spatiotemporal adaptive compression for long video-language understanding","year":"2024","author":"Shen","key":"ref69"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.393"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.01725"},{"article-title":"Llama: Open and efficient foundation language models","year":"2023","author":"Touvron","key":"ref72"},{"article-title":"Llama 2: Open foundation and fine-tuned chat models","year":"2023","author":"Touvron","key":"ref73"},{"article-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution","year":"2024","author":"Wang","key":"ref74"},{"article-title":"Internvid: A large-scale videotext dataset for multimodal understanding and generation","year":"2023","author":"Wang","key":"ref75"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.32388\/xngycy"},{"article-title":"Videollamb: Long-context video understanding with recurrent memory bridges","year":"2024","author":"Wang","key":"ref77"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00965"},{"article-title":"Streaming video understanding and multi-round interaction with memory-enhanced knowledge","volume-title":"ICLR","author":"Xiong","key":"ref79"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00498"},{"article-title":"Vidchapters-7m: Video chapters at scale","volume-title":"NeurIPS","author":"Yang","key":"ref81"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01032"},{"article-title":"Qwen2 technical report","year":"2024","author":"Yang","key":"ref83"},{"article-title":"Deepspeedvisualchat: Multi-round multi-image interleave chat via multi-modal causal attention","year":"2023","author":"Yao","key":"ref84"},{"article-title":"Ferret: Refer and ground anything anywhere at any granularity","year":"2023","author":"You","key":"ref85"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01589"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"},{"author":"Zhang","key":"ref88","article-title":"Videollama 3: Frontier multimodal foundation models for image and video understanding"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-demo.49"},{"article-title":"Flash-vstream: Memorybased real-time understanding for long video streams","year":"2024","author":"Zhang","key":"ref90"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-naacl.51"},{"article-title":"Internlm-xcomposer-2.5: A versatile large vision language model supporting long-contextual input and output","year":"2024","author":"Zhang","key":"ref92"},{"article-title":"Long context transfer from language to vision","year":"2024","author":"Zhang","key":"ref93"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-91813-1_4"},{"article-title":"Video instruction tuning with synthetic data","year":"2024","author":"Zhang","key":"ref95"},{"article-title":"Video instruction tuning with synthetic data","year":"2024","author":"Zhang","key":"ref96"},{"key":"ref97","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19830-4_28"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01403"},{"article-title":"Judging llm-as-a-judge with mt-bench and chatbot arena","volume-title":"NeurIPS","author":"Zheng","key":"ref99"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.12342"},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01727"},{"article-title":"Minigpt-4: Enhancing vision-language understanding with advanced large language models","year":"2023","author":"Zhu","key":"ref102"}],"event":{"name":"2025 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","start":{"date-parts":[[2025,6,10]]},"location":"Nashville, TN, USA","end":{"date-parts":[[2025,6,17]]}},"container-title":["2025 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11091818\/11091608\/11092811.pdf?arnumber=11092811","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,15]],"date-time":"2025-08-15T04:47:58Z","timestamp":1755233278000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11092811\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,10]]},"references-count":102,"URL":"https:\/\/doi.org\/10.1109\/cvpr52734.2025.02708","relation":{},"subject":[],"published":{"date-parts":[[2025,6,10]]}}}