{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T20:22:30Z","timestamp":1792786950831,"version":"4.5.0"},"reference-count":160,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"10","license":[{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Natural Science Foundation of China","award":["62372223"],"award-info":[{"award-number":["62372223"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"crossref"}]},{"name":"National Key R&D Program of China","award":["2022ZD0161300"],"award-info":[{"award-number":["2022ZD0161300"]}]},{"name":"Key R&D Program of Xinjiang Uygur Autonomous Region"},{"name":"Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China","award":["JYB2025XDXM118"],"award-info":[{"award-number":["JYB2025XDXM118"]}],"id":[{"id":"https:\/\/ror.org\/03m01yf64","id-type":"ROR","asserted-by":"crossref"}]},{"name":"National Natural Science Foundation of China","award":["623B2050"],"award-info":[{"award-number":["623B2050"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,10]]},"DOI":"10.1109\/tpami.2026.3702168","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T20:02:10Z","timestamp":1781121730000},"page":"12582-12600","source":"Crossref","is-referenced-by-count":0,"title":["Revisiting InternVL: A Systematic Technical Framework for Building Powerful Open-Source Vision-Language Models"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3419-8812","authenticated-orcid":false,"given":"Zhe","family":"Chen","sequence":"first","affiliation":[{"name":"Nanjing University","place":["Nanjing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weiyun","family":"Wang","sequence":"additional","affiliation":[{"name":"Fudan University","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinguo","family":"Zhu","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8402-7504","authenticated-orcid":false,"given":"Wenhai","family":"Wang","sequence":"additional","affiliation":[{"name":"The Chinese University of Hong Kong","place":["Shatin, Hong Kong"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yangzhou","family":"Liu","sequence":"additional","affiliation":[{"name":"Nanjing University","place":["Nanjing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhaoyang","family":"Liu","sequence":"additional","affiliation":[{"name":"The Hong Kong University of Science and Technology","place":["Clear Water Bay, Hong Kong"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gen","family":"Luo","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peng","family":"Sun","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weijie","family":"Su","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-0941-3629","authenticated-orcid":false,"given":"Hao","family":"Tian","sequence":"additional","affiliation":[{"name":"SenseTime Research","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-9809-3818","authenticated-orcid":false,"given":"Lewei","family":"Lu","sequence":"additional","affiliation":[{"name":"SenseTime Research","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xizhou","family":"Zhu","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7051-5347","authenticated-orcid":false,"given":"Tong","family":"Lu","sequence":"additional","affiliation":[{"name":"Nanjing University","place":["Nanjing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dahua","family":"Lin","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1889-2567","authenticated-orcid":false,"given":"Yu","family":"Qiao","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6785-0785","authenticated-orcid":false,"given":"Jifeng","family":"Dai","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"GPT-4V(ision) system card","year":"2023"},{"key":"ref2","article-title":"GPT-4O system card","author":"Hurst","year":"2024"},{"key":"ref3","article-title":"Introducing GPT-5","year":"2025"},{"key":"ref4","article-title":"Gemini: A family of highly capable multimodal models","author":"Team","year":"2023"},{"key":"ref5","article-title":"Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context","author":"Reid","year":"2024"},{"key":"ref6","article-title":"Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities","author":"Comanici","year":"2025"},{"key":"ref7","article-title":"Gemini 3 pro - model card","year":"2025"},{"key":"ref8","article-title":"Qwen-Vl: A frontier large vision-language model with versatile abilities","author":"Bai","year":"2023"},{"key":"ref9","article-title":"Qwen2-Vl: Enhancing vision-language model\u2019s perception of the world at any resolution","author":"Wang","year":"2024"},{"key":"ref10","article-title":"Qwen2.5-Vl technical report","author":"Bai","year":"2025"},{"key":"ref11","article-title":"Qwen3-Vl technical report","author":"Bai","year":"2025"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02283"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4231-5"},{"key":"ref14","article-title":"Expanding performance boundaries of open-source multimodal models with model, data, and test-time scaling","author":"Chen","year":"2024"},{"key":"ref15","article-title":"InternVL3: Exploring advanced training and test-time recipes for open-source multimodal models","author":"Zhu","year":"2025"},{"key":"ref16","first-page":"92628","article-title":"Eagle: Exploring the design space for multimodal LLMs with mixture of encoders","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Shi"},{"key":"ref17","article-title":"Eagle 2: Building post-training data strategies from scratch for frontier vision-language models","author":"Li","year":"2025"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.52202\/085713-3047"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3860"},{"key":"ref20","article-title":"CogVLM2: Visual language models for image and video understanding","author":"Hong","year":"2024"},{"key":"ref21","article-title":"MiniCPM-V: A GPT-4V level MLLM on your phone","author":"Yao","year":"2024"},{"key":"ref22","article-title":"MiniCPM-V4.5: Cooking efficient MLLMs via architecture, data, and training recipe","author":"Yu","year":"2025"},{"key":"ref23","article-title":"Deepseek-Vl: Towards real-world vision-language understanding","author":"Lu","year":"2024"},{"key":"ref24","article-title":"Deepseek-Vl2: Mixture-of-experts vision-language models for advanced multimodal understanding","author":"Wu","year":"2024"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref27","article-title":"LLaVA-onevision: Easy visual task transfer","volume":"2025","author":"Li","year":"2025","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref28","article-title":"LLaVA-onevision-1.5: Fully open framework for democratized multimodal training","author":"An","year":"2025"},{"key":"ref29","article-title":"InternLM2 technical report","author":"Cai","year":"2024"},{"key":"ref30","article-title":"Qwen2 technical report","author":"Team","year":"2024"},{"key":"ref31","article-title":"Qwen3 technical report","author":"Yang","year":"2025"},{"key":"ref32","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Radford"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02520"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00390"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"ref37","first-page":"1","article-title":"BEiT: Bert pre-training of image transformers","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Bao"},{"key":"ref38","article-title":"BEiT V2: Masked image modeling with vector-quantized visual tokenizers","author":"Peng","year":"2022"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01838"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01855"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1016\/j.imavis.2024.105171"},{"key":"ref42","article-title":"DINOv2: Learning robust visual features without supervision","volume":"2024","author":"Oquab","year":"2024","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref43","article-title":"EVA-clip: Improved training techniques for clip at scale","author":"Sun","year":"2023"},{"key":"ref44","first-page":"304","article-title":"MM1: Methods, analysis and insights from multimodal LLM pre-training","volume-title":"Proc. Eur. Conf. Comput. Vis.","author":"McKinzie"},{"key":"ref45","first-page":"64669","article-title":"MM1.5: Methods, analysis & insights from multimodal LLM fine-tuning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zhang"},{"key":"ref46","article-title":"Molmo and PixMo: Open weights and open data for state-of-the-art multimodal models","author":"Deitke","year":"2024"},{"key":"ref47","article-title":"SigLIP 2: Multilingual vision-language encoders with improved semantic understanding, localization, and dense features","author":"Tschannen","year":"2025"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00901"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.52202\/075280-0106"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.52202\/075280-3155"},{"key":"ref51","article-title":"Introducing our multimodal models","author":"Bavishi","year":"2023"},{"key":"ref52","article-title":"The Llama 4 herd: Architecture, training, evaluation, and deployment notes","author":"Adcock","year":"2026"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2010.11929"},{"key":"ref54","first-page":"7480","article-title":"Scaling vision transformers to 22 billion parameters","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Dehghani"},{"key":"ref55","article-title":"Root mean square layer normalization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhang"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1833"},{"key":"ref57","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023"},{"key":"ref58","article-title":"Laion COCO: 600M synthetic captions from laion2B-en","author":"Schuhmann","year":"2022"},{"key":"ref59","article-title":"Coyo-700M: Image-text pair dataset","author":"Byeon","year":"2022"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1916"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00356"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P18-1238"},{"key":"ref63","first-page":"1143","article-title":"Im2Text: Describing images using 1 million captioned photographs","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ordonez"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02240"},{"key":"ref65","article-title":"PP-OCRv3: More attempts for the improvement of ultra lightweight OCR system","author":"Li","year":"2022"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1007\/s44267-024-00067-6"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01958"},{"key":"ref68","article-title":"Phi-3 technical report: A highly capable language model locally on your phone","author":"Abdin","year":"2024"},{"key":"ref69","article-title":"YI: Open foundation models by 01.AI","author":"Young","year":"2024"},{"key":"ref70","article-title":"The Llama 3 herd of models","author":"Dubey","year":"2024"},{"key":"ref71","article-title":"SmoLLM-corpus","author":"Allal"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.206"},{"key":"ref73","article-title":"SCP-116K: A high-quality problem-solution dataset and a generalized pipeline for automated extraction in the higher education science domain","author":"Lu","year":"2025"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00276"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00505"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref77","article-title":"Are we done with imagenet?","author":"Beyer","year":"2020"},{"key":"ref78","first-page":"5389","article-title":"Do imageNet classifiers generalize to imagenet?","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Recht"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01501"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00823"},{"key":"ref81","first-page":"10506","article-title":"Learning robust global representations by penalizing local predictive power","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wang"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01228-1_26"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.544"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00132"},{"key":"ref85","article-title":"Enhancing the reasoning ability of multimodal large language models via mixed preference optimization","author":"Wang","year":"2024"},{"key":"ref86","article-title":"Deepseekmath: Pushing the limits of mathematical reasoning in open language models","author":"Shao","year":"2024"},{"key":"ref87","first-page":"216","article-title":"MBBench: Is your multi-modal model an all-around player?","volume-title":"Proc. Eur. Conf. Comput. Vis.","author":"Liu"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0850"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00913"},{"key":"ref90","first-page":"23439","article-title":"MathVista: Evaluating mathematical reasoning of foundation models in visual contexts","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Lu"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01363"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_15"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4235-6"},{"key":"ref94","first-page":"57730","article-title":"MM-Vet: Evaluating large multimodal models for integrated capabilities","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Yu"},{"key":"ref95","article-title":"Infinity-MM: Scaling multimodal performance with large-scale and high-quality instruction data","author":"Gu","year":"2024"},{"key":"ref96","article-title":"Ovis: Structural embedding alignment for multimodal large language model","author":"Lu","year":"2024"},{"key":"ref97","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2771"},{"key":"ref98","article-title":"The claude 3 model family: Opus, sonnet, haiku","year":"2024"},{"key":"ref99","article-title":"NVLM: Open frontier-class multimodal LLMs","author":"Dai","year":"2024"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-acl.177"},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1109\/WACV48630.2021.00225"},{"key":"ref103","doi-asserted-by":"publisher","DOI":"10.1109\/WACV51458.2022.00264"},{"key":"ref104","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.01263"},{"key":"ref105","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3609"},{"key":"ref106","article-title":"VCR: Visual caption restoration","author":"Zhang","year":"2024"},{"key":"ref107","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73337-6_9"},{"key":"ref108","article-title":"MANTIS: Interleaved multi-image instruction tuning","volume":"2024","author":"Jiang","year":"2024","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref109","first-page":"38405","article-title":"MMIU: Multimodal multi-image understanding for evaluating large vision-language models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Meng"},{"key":"ref110","first-page":"62624","article-title":"MuirBench: A comprehensive benchmark for robust multi-image understanding","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Wang"},{"key":"ref111","first-page":"57116","article-title":"MMT-bench: A comprehensive multimodal benchmark for evaluating large vision-language models towards multitask AGI","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Ying"},{"key":"ref112","article-title":"Benchmarking multi-image understanding in vision and language models: Perception, knowledge, reasoning, and multi-hop reasoning","author":"Zhao","year":"2024"},{"key":"ref113","article-title":"Grok-1.5 vision preview","author":"AI","year":"2024"},{"key":"ref114","article-title":"MME-realworld: Could your multimodal LLM challenge high-resolution real-world scenarios that are difficult for humans?","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zhang"},{"key":"ref115","doi-asserted-by":"publisher","DOI":"10.52202\/079017-1528"},{"key":"ref116","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2025.3558652"},{"key":"ref117","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1086"},{"key":"ref118","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.9"},{"key":"ref119","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72970-6_3"},{"key":"ref120","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01471"},{"key":"ref121","article-title":"One-peace: Exploring one general representation model toward unlimited modalities","author":"Wang","year":"2023"},{"key":"ref122","article-title":"Shikra: Unleashing multimodal LLM\u2019s referential dialogue magic","author":"Chen","year":"2023"},{"key":"ref123","article-title":"Ferret-V2: An improved baseline for referring and grounding with large language models","author":"Zhang","year":"2024"},{"key":"ref124","article-title":"TextHawk2: A large vision-language model excels in bilingual ocr and grounding with 16x fewer tokens","author":"Yu","year":"2024"},{"key":"ref125","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4321-9"},{"key":"ref126","first-page":"85485","article-title":"Oryx MLLM: On-demand spatial-temporal understanding at arbitrary resolution","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Liu"},{"key":"ref127","article-title":"VideoLLaMA 2: Advancing spatial-temporal modeling and audio understanding in video-LLMs","author":"Cheng","year":"2024"},{"key":"ref128","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02245"},{"key":"ref129","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02095"},{"key":"ref130","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2827"},{"key":"ref131","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01278"},{"key":"ref132","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0907"},{"key":"ref133","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1189"},{"key":"ref134","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00380"},{"key":"ref135","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref136","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0182"},{"key":"ref137","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref138","article-title":"Tiny LVLM-ehub: Early multimodal experiments with bard","author":"Shao","year":"2023"},{"key":"ref139","article-title":"Seed-bench: Benchmarking multimodal LLMs with generative comprehension","author":"Li","year":"2023"},{"key":"ref140","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3014"},{"key":"ref141","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73242-3_10"},{"key":"ref142","first-page":"48337","article-title":"DynaMath: A dynamic visual benchmark for evaluating mathematical reasoning robustness of vision language models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zou"},{"key":"ref143","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.983"},{"key":"ref144","article-title":"LogicVista: Multimodal LLM logical reasoning benchmark in visual contexts","author":"Xiao","year":"2024"},{"key":"ref145","article-title":"Measuring massive multitask language understanding","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Hendrycks"},{"key":"ref146","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.671"},{"key":"ref147","doi-asserted-by":"crossref","DOI":"10.52202\/075280-2749","article-title":"C-EVAL: A multi-level multi-discipline chinese evaluation suite for foundation models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Huang"},{"key":"ref148","article-title":"Evaluating the performance of large language models on gaokao benchmark","author":"Zhang","year":"2023"},{"key":"ref149","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P17-1147"},{"key":"ref150","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00276"},{"key":"ref151","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00305"},{"key":"ref152","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1082"},{"key":"ref153","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6399"},{"key":"ref154","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/p19-1472"},{"key":"ref155","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-acl.824"},{"key":"ref156","article-title":"Training verifiers to solve math word problems","author":"Cobbe","year":"2021"},{"issue":"4","key":"ref157","first-page":"1","article-title":"Measuring mathematical problem solving with the MATH dataset","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"2","author":"Hendrycks","year":"2021"},{"key":"ref158","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.489"},{"key":"ref159","article-title":"Evaluating large language models trained on code","author":"Chen","year":"2021"},{"key":"ref160","article-title":"Program synthesis with large language models","author":"Austin","year":"2021"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11674301\/11556915.pdf?arnumber=11556915","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T16:58:40Z","timestamp":1790269120000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11556915\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10]]},"references-count":160,"journal-issue":{"issue":"10"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2026.3702168","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10]]}}}