{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T16:25:10Z","timestamp":1778171110326,"version":"3.51.4"},"reference-count":100,"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\/iccv51701.2025.01641","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"17663-17674","source":"Crossref","is-referenced-by-count":1,"title":["Are They the Same? Exploring Visual Correspondence Shortcomings of Multimodal LLMs"],"prefix":"10.1109","author":[{"given":"Yikang","family":"Zhou","sequence":"first","affiliation":[{"name":"Wuhan University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Wuhan University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shilin","family":"Xu","sequence":"additional","affiliation":[{"name":"Peking University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shihao","family":"Chen","sequence":"additional","affiliation":[{"name":"Wuhan University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qianyu","family":"Zhou","sequence":"additional","affiliation":[{"name":"SJTU"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yunhai","family":"Tong","sequence":"additional","affiliation":[{"name":"Peking University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shunping","family":"Ji","sequence":"additional","affiliation":[{"name":"Wuhan University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiangning","family":"Zhang","sequence":"additional","affiliation":[{"name":"Zhejiang University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lu","family":"Qi","sequence":"additional","affiliation":[{"name":"Wuhan University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiangtai","family":"Li","sequence":"additional","affiliation":[{"name":"Bytedance Seed"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","author":"Abdin","year":"2024","journal-title":"Phi-3 technical report: A highly capable language model locally on your phone"},{"key":"ref2","author":"Anil","year":"2023","journal-title":"Palm 2 technical report"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/WACV56688.2023.00172"},{"key":"ref4","author":"Bai","year":"2023","journal-title":"Qwen technical report"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2000.855899"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01227"},{"key":"ref7","article-title":"The wildtrack multi-camera person dataset","author":"Chavdarova","year":"2018","journal-title":"CVPR"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72643-9_22"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0850"},{"key":"ref10","author":"Chen","year":"2020","journal-title":"Improved baselines with momentum contrastive learning"},{"key":"ref11","author":"Chen","year":"2024","journal-title":"Expanding performance boundaries of open-source multimodal models with model, data, and test-time scaling"},{"key":"ref12","article-title":"Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks","author":"Chen","year":"2024","journal-title":"CVPR"},{"key":"ref13","volume-title":"XTuner Contributors. Xtuner: A toolkit for efficiently fine-tuning llm","year":"2023"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00910"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2142"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01850"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00552"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3301975"},{"key":"ref19","author":"Fu","year":"2024","journal-title":"Mme: A comprehensive evaluation benchmark for multimodal large language models"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73337-6_9"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/IVS.2011.5940405"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.670"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00975"},{"key":"ref24","article-title":"3d-llm: Injecting the 3d world into large language models","author":"Hong","year":"2023","journal-title":"NeurIPS"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58571-6_1"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3153312"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref28","author":"Touvron","year":"2023","journal-title":"Llama 2: Open foundation and fine-tuned chat models"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref30","author":"Lai","year":"2023","journal-title":"Lisa: Reasoning segmentation via large language model"},{"key":"ref31","author":"Li","year":"2024","journal-title":"Naturalbench: Evaluating vision-language models on natural adversarial samples"},{"key":"ref32","author":"Li","year":"2024","journal-title":"Llava-onevision: Easy visual task transfer"},{"key":"ref33","author":"Li","year":"2024","journal-title":"Llava-next-interleave: Tackling multi-image, video, and 3d in large multimodal models"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01794"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01828"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01280"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref38","article-title":"Llama-vid: An image is worth 2 tokens in large language models","author":"Li","year":"2024","journal-title":"ECCV"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.342"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02520"},{"key":"ref41","author":"Lin","year":"2023","journal-title":"Mm-vid: Advancing video understanding with gpt-4v (ision)"},{"key":"ref42","author":"Lin","year":"2024","journal-title":"Draw-and-understand: Leveraging visual prompts to enable mllms to comprehend what you want"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02484"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref46","author":"Liu","year":"2024","journal-title":"Llava-next: Improved reasoning, ocr, and world knowledge"},{"key":"ref47","article-title":"Mmbench: Is your multimodal model an all-around player?","author":"Liu","year":"2024","journal-title":"ECCV"},{"key":"ref48","author":"Lu","year":"2024","journal-title":"Deepseek-vl: towards real-world visionlanguage understanding"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.679"},{"key":"ref50","author":"Meng","year":"2024","journal-title":"Mmiu: Multimodal multi-image understanding for evaluating large vision-language models"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.02036"},{"key":"ref52","author":"Milan","year":"2016","journal-title":"Mot16: A benchmark for multi-object tracking"},{"key":"ref53","year":"2023","journal-title":"OpenAI. Gpt-4 technical report"},{"key":"ref54","author":"Oquab","year":"2023","journal-title":"Dinov2: Learning robust visual features without supervision"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.52202\/079017-4157"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-022-01629-1"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW63382.2024.00191"},{"key":"ref58","article-title":"Learning transferable visual models from natural language supervision","author":"Radford","year":"2021","journal-title":"ICML"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01187"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01236"},{"key":"ref61","author":"Ravi","year":"2024","journal-title":"Sam 2: Segment anything in images and videos"},{"key":"ref62","author":"Shi","year":"2024","journal-title":"Eagle: Exploring the design space for multimodal 11 ms with mixture of encoders"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01725"},{"key":"ref65","author":"Sun","year":"2024","journal-title":"Chattracker: Enhancing visual tracking performance via chatting with multimodal large language model"},{"key":"ref66","year":"2023","journal-title":"InternLM Team. Internlm: A multilingual language model with progressively enhanced capabilities"},{"key":"ref67","author":"Tong","year":"2024","journal-title":"Cambrian1: A fully open, vision-centric exploration of multimodal llms"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00914"},{"key":"ref69","author":"Touvron","year":"2023","journal-title":"Llama: Open and efficient foundation language models"},{"key":"ref70","author":"Touvron","year":"2023","journal-title":"Llama: Open and efficient foundation language models"},{"key":"ref71","author":"Touvron","year":"2023","journal-title":"Llama 2: Open foundation and fine-tuned chat models"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01219-9_41"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-emnlp.50"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-024-02076-w"},{"key":"ref75","author":"Wang","year":"2024","journal-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution"},{"key":"ref76","author":"Wang","year":"2023","journal-title":"Chat-3d: Data-efficiently tuning large language model for universal dialogue of 3d scenes"},{"key":"ref77","article-title":"Chain-of-thought prompting elicits reasoning in large language models","author":"Wei","year":"2022","journal-title":"NeurIPS"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01548"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19815-1_34"},{"key":"ref80","author":"Wu","year":"2024","journal-title":"Controlmllm: Training-free visual prompt learning for multimodal large language models"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72698-9_8"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19803-8_43"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01471"},{"key":"ref84","author":"Yang","year":"2024","journal-title":"Qwen2 technical report"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00529"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00089"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00271"},{"key":"ref88","author":"Yu","year":"2023","journal-title":"Mm -vet: Evaluating large multimodal models for integrated capabilities"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02664"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.01767"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00913"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-demo.49"},{"key":"ref93","author":"Zhang","year":"2023","journal-title":"1st place solution for pvuw challenge 2023: Video panoptic segmentation"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00124"},{"key":"ref95","author":"Zhang","year":"2023","journal-title":"Dvis++: Improved decoupled framework for universal video segmentation"},{"key":"ref96","author":"Zhang","year":"2023","journal-title":"1st place solution for the 5th lsvos challenge: Video instance segmentation"},{"key":"ref97","article-title":"Omg-llava: Bridging image-level, object-level, pixel-level reasoning and understanding","author":"Zhang","year":"2024","journal-title":"NeurIPS"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00637"},{"key":"ref99","article-title":"Dvis-daq: Improving video segmentation via dynamic anchor queries","author":"Zhou","year":"2024","journal-title":"ECCV"},{"key":"ref100","author":"Zolkepli","year":"2024","journal-title":"Mmmmodal - multi-images multi-audio multiturn multi-modal"}],"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\/11445726.pdf?arnumber=11445726","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:17:24Z","timestamp":1777612644000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11445726\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":100,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01641","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}