{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,29]],"date-time":"2026-05-29T12:34:20Z","timestamp":1780058060291,"version":"3.54.0"},"reference-count":84,"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"}],"funder":[{"DOI":"10.13039\/501100003452","name":"Innovation and Technology Commission","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003452","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.00352","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"3694-3704","source":"Crossref","is-referenced-by-count":0,"title":["LYRA: An Efficient and Speech-Centric Framework for Omni-Cognition"],"prefix":"10.1109","author":[{"given":"Zhisheng","family":"Zhong","sequence":"first","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chengyao","family":"Wang","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuqi","family":"Liu","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Senqiao","family":"Yang","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Longxiang","family":"Tang","sequence":"additional","affiliation":[{"name":"HKUST"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuechen","family":"Zhang","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingyao","family":"Li","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tianyuan","family":"Qu","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanwei","family":"Li","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yukang","family":"Chen","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shaozuo","family":"Yu","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sitong","family":"Wu","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eric","family":"Lo","sequence":"additional","affiliation":[{"name":"CUHK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shu","family":"Liu","sequence":"additional","affiliation":[{"name":"SmartMore"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiaya","family":"Jia","sequence":"additional","affiliation":[{"name":"HKUST"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","volume-title":"ChatTTS","year":"2024"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1723"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46454-1_24"},{"key":"ref4","author":"Bai","year":"2023","journal-title":"Qwen-VL: A frontier large vision-language model with versatile abilities"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.172"},{"key":"ref6","article-title":"AuroraCap: Efficient, performant video detailed captioning and a new benchmark","author":"Chai","year":"2025","journal-title":"ICLR"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.00513"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73004-7_2"},{"key":"ref9","article-title":"LongLoRA: Efficient fine-tuning of long-context large language models","author":"Chen","year":"2024","journal-title":"ICLR"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4231-5"},{"key":"ref11","article-title":"Qwen2-audio technical report","author":"Chu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2142"},{"key":"ref13","article-title":"High fidelity neural audio compression","author":"D\u00e9fossez","journal-title":"Transactions on Machine Learning Research"},{"key":"ref14","article-title":"Moshi: a speech-text foundation model for realtime dialogue","author":"D\u00e9fossez","year":"2024","journal-title":"arXiv preprint"},{"key":"ref15","article-title":"The LLaMA 3 herd of models","author":"Dubey","year":"2024","journal-title":"arXiv preprint"},{"key":"ref16","article-title":"LlaMA- Omni: Seamless speech interaction with large language models","author":"Fang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref17","article-title":"MME: A comprehensive evaluation benchmark for multimodal large language models","author":"Fu","year":"2023","journal-title":"2306"},{"key":"ref18","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":"ref19","article-title":"VITA: Towards open-source interactive omni multimodal LLM","author":"Fu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref20","article-title":"Model tells you what to discard: Adaptive KV cache compression for Llms","author":"Ge","year":"2023","journal-title":"arXiv preprint"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01457"},{"key":"ref22","article-title":"Automated audio captioning by fine-tuning bart with audioset tags","volume-title":"DCASE 2021\u20136th Workshop on Detection and Classification of Acoustic Scenes and Events","author":"Gontier","year":"2021"},{"key":"ref23","volume-title":"Gemma: Introducing new state-of-the-art open models","year":"2024"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref26","article-title":"LoRA: Low-rank adaptation of large language models","author":"Hu","year":"2021","journal-title":"ICLR"},{"key":"ref27","article-title":"Scaling up visual and vision-language representation learning with noisy text supervision","author":"Jia","year":"2021","journal-title":"ICML"},{"key":"ref28","author":"Albert","year":"2024","journal-title":"Mixtral of experts"},{"key":"ref29","first-page":"119","article-title":"AudioCaps: Generating captions for audios in the wild","volume-title":"Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies","volume":"1","author":"Kim","year":"2019"},{"key":"ref30","article-title":"Exploring train and test-time augmentations for audio-language learning","author":"Kim","year":"2022","journal-title":"arXiv preprint"},{"key":"ref31","article-title":"LLaVA- OneVision: Easy visual task transfer","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref32","first-page":"19730","article-title":"BLIP-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","author":"Li","year":"2023","journal-title":"ICML"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02095"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3637265"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_19"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02520"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref39","author":"Liu","year":"2024","journal-title":"LLaVA- NeXT: Improved reasoning, ocr, and world knowledge"},{"key":"ref40","article-title":"Scissorhands: Exploiting the persistence of importance hypothesis for LLM KV cache compression at test time","volume-title":"NeruIPS, 36","author":"Liu","year":"2024"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.85"},{"key":"ref42","article-title":"Egoschema: A diagnostic benchmark for very longform video language understanding","volume":"36","author":"Mangalam","year":"2023","journal-title":"NeurIPS"},{"key":"ref43","article-title":"Audio captioning transformer","author":"Mei","year":"2021","journal-title":"arXiv preprint"},{"key":"ref44","article-title":"Microsoft","volume-title":"Edge-TTS","year":"2024"},{"key":"ref45","volume-title":"ChatGPT","year":"2023"},{"key":"ref46","article-title":"OpenAI","volume-title":"GPT-4o","year":"2024"},{"key":"ref47","first-page":"2024","year":"2024","journal-title":"InternOmni: Extending internvl with audio modality"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-475"},{"key":"ref50","article-title":"Learning transferable visual models from natural language supervision","author":"Radford","year":"2021","journal-title":"ICML"},{"key":"ref51","first-page":"28492","article-title":"Robust speech recognition via large-scale weak supervision","author":"Radford","year":"2023","journal-title":"ICML"},{"key":"ref52","article-title":"AudioPaLM: A large language model that can speak and listen","author":"Paul","year":"2023","journal-title":"arXiv preprint"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2005.11.008"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref55","article-title":"Snac: Multi-scale neural audio codec","author":"Siuzdak","year":"2024","journal-title":"arXiv preprint"},{"key":"ref56","article-title":"SALMONN: Towards generic hearing abilities for large language models","author":"Tang","year":"2024","journal-title":"ICLR"},{"key":"ref57","article-title":"Quest: Query-aware sparsity for efficient long-context 11 m inference","author":"Tang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref58","article-title":"Any-to-any generation via composable diffusion","author":"Tang","year":"2024","journal-title":"NeurIPS, 36"},{"key":"ref59","volume-title":"Stanford Alpaca: An instruction-following llama model","author":"Taori","year":"2023"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-86331-9_50"},{"key":"ref61","author":"Touvron","year":"2023","journal-title":"LLaMA: Open and efficient foundation language models"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7299087"},{"key":"ref63","volume-title":"Common Voice","author":"Voice","year":"2024"},{"key":"ref64","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":"ref65","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3860"},{"key":"ref66","article-title":"Longvideobench: A benchmark for long-context interleaved video-language understanding","author":"Wu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref67","article-title":"Next-GPT: Any-to-any multimodal 1lm","author":"Wu","year":"2023","journal-title":"arXiv preprint"},{"key":"ref68","article-title":"Efficient streaming language models with attention sinks","author":"Xiao","year":"2023","journal-title":"arXiv preprint"},{"key":"ref69","article-title":"Mini-Omni: Language models can hear, talk while thinking in streaming","author":"Xie","year":"2024","journal-title":"arXiv preprint"},{"key":"ref70","article-title":"LongVILA: Scaling long-context visual language models for long videos","author":"Xue","year":"2024","journal-title":"arXiv preprint"},{"key":"ref71","article-title":"Qwen2 technical report","author":"Yang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.109"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01843"},{"key":"ref74","article-title":"Minicpm-v: A gpt-4v level mllm on your phone","author":"Yao","year":"2024","journal-title":"arXiv preprint"},{"key":"ref75","author":"Yu","year":"2023","journal-title":"MM- Vet: Evaluating large multimodal models for integrated capabilities"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00913"},{"key":"ref77","article-title":"AnyGPT: Unified multimodal 11 m with discrete sequence modeling","author":"Zhan","year":"2024","journal-title":"arXiv preprint"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.1055"},{"key":"ref79","article-title":"Internlm-xcomposer2. 5-omnilive: A comprehensive multimodal system for long-term streaming video and audio interactions","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref80","article-title":"Long context transfer from language to vision","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.485"},{"key":"ref82","article-title":"Speechtokenizer: Unified speech tokenizer for speech language models","author":"Zhang","year":"2024","journal-title":"ICLR"},{"key":"ref83","first-page":"34661","article-title":"H2O: Heavy-hitter oracle for efficient generative inference of large language models","volume":"36","author":"Zhang","year":"2023","journal-title":"NeruIPS"},{"key":"ref84","author":"Zhu","year":"2023","journal-title":"MiniGPT-4: Enhancing visionlanguage understanding with advanced large language models"}],"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\/11444752.pdf?arnumber=11444752","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T06:46:28Z","timestamp":1777531588000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444752\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":84,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.00352","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}