{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T08:16:21Z","timestamp":1783152981999,"version":"3.54.6"},"publisher-location":"New York, NY, USA","reference-count":43,"publisher":"ACM","funder":[{"name":"National Natural Science Foundation of China","award":["62402050"],"award-info":[{"award-number":["62402050"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,4,13]]},"DOI":"10.1145\/3774904.3792127","type":"proceedings-article","created":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T21:54:34Z","timestamp":1775771674000},"page":"5021-5030","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Task-Aware Cloud-End Offloading for Vision-Language Model Serving via Dynamic Modality-Specific Adapter Scheduling"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-6605-0920","authenticated-orcid":false,"given":"Zian","family":"Wang","sequence":"first","affiliation":[{"name":"Beijing University of Posts and Telecommunications, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8174-0593","authenticated-orcid":false,"given":"Ziyi","family":"Wang","sequence":"additional","affiliation":[{"name":"Beijing University of Posts and Telecommunications, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-9747-8539","authenticated-orcid":false,"given":"Jie","family":"Xing","sequence":"additional","affiliation":[{"name":"Beijing University of Posts and Telecommunications, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-7117-4662","authenticated-orcid":false,"given":"Yaya","family":"Wei","sequence":"additional","affiliation":[{"name":"Omni-Channel Operation Center, China Telecommunications Corporation, 0009-0002-9747-8539, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-7362-8999","authenticated-orcid":false,"given":"Ziyan","family":"Zhong","sequence":"additional","affiliation":[{"name":"Omni-Channel Operation Center, China Telecommunications Corporation, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0674-7864","authenticated-orcid":false,"given":"Lanshan","family":"Zhang","sequence":"additional","affiliation":[{"name":"Beijing University of Posts and Telecommunications, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,4,12]]},"reference":[{"key":"e_1_3_2_1_1_1","volume-title":"Mega-bench: Scaling multimodal evaluation to over 500 real-world tasks. arXiv preprint arXiv:2410.10563","author":"Chen J.","year":"2024","unstructured":"Chen, J., Liang, T., Siu, S., Wang, Z., Wang, K., Wang, Y., Ni, Y., Zhu, W., Jiang, Z., Lyu, B., et al. Mega-bench: Scaling multimodal evaluation to over 500 real-world tasks. arXiv preprint arXiv:2410.10563 (2024)."},{"key":"e_1_3_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02283"},{"key":"e_1_3_2_1_3_1","volume-title":"Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks","author":"Chen Z.","year":"2024","unstructured":"Chen, Z., Wu, J., Wang, W., Su, W., Chen, G., Xing, S., Zhong, M., Zhang, Q., Zhu, X., Lu, L., Li, B., Luo, P., Lu, T., Qiao, Y., and Dai, J. Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks, 2024."},{"key":"e_1_3_2_1_4_1","volume-title":"Qlora: Efficient finetuning of quantized llms. Advances in neural information processing systems 36","author":"Dettmers T.","year":"2023","unstructured":"Dettmers, T., Pagnoni, A., Holtzman, A., and Zettlemoyer, L. Qlora: Efficient finetuning of quantized llms. Advances in neural information processing systems 36 (2023), 10088--10115."},{"key":"e_1_3_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1145\/3543507.3583540"},{"key":"e_1_3_2_1_6_1","volume-title":"Inference economics of language models. arXiv preprint arXiv:2506.04645","author":"Erdil E.","year":"2025","unstructured":"Erdil, E. Inference economics of language models. arXiv preprint arXiv:2506.04645 (2025)."},{"key":"e_1_3_2_1_7_1","volume-title":"Accelerating end-cloud collaborative inference via near bubble-free pipeline optimization","author":"Gao L.","year":"2024","unstructured":"Gao, L., Liu, J., Xu, H., Xu, S., Ma, Q., and Huang, L. Accelerating end-cloud collaborative inference via near bubble-free pipeline optimization, 2024."},{"key":"e_1_3_2_1_8_1","volume-title":"Hugging Face Blog","author":"Gigant T.","year":"2024","unstructured":"Gigant, T. Design choices for vision language models in 2024. Hugging Face Blog, 2024. [Online]. Available: https:\/\/huggingface.co\/blog\/vision-language-models."},{"key":"e_1_3_2_1_9_1","volume-title":"Mamba: Linear-time sequence modeling with selective state spaces","author":"Gu A.","year":"2024","unstructured":"Gu, A., and Dao, T. Mamba: Linear-time sequence modeling with selective state spaces, 2024."},{"key":"e_1_3_2_1_10_1","volume-title":"Efficient deployment of visionlanguage models on mobile devices: A case study on oneplus 13r. arXiv preprint arXiv:2507.08505","author":"Guerrero P. R.","year":"2025","unstructured":"Guerrero, P. R., Pan, Y., and Kashyap, S. Efficient deployment of visionlanguage models on mobile devices: A case study on oneplus 13r. arXiv preprint arXiv:2507.08505 (2025)."},{"key":"e_1_3_2_1_11_1","volume-title":"International Conference on Learning Representations","author":"Hu E. J.","year":"2022","unstructured":"Hu, E. J., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., Chen, W., et al. Lora: Low-rank adaptation of large language models. In International Conference on Learning Representations (2022)."},{"key":"e_1_3_2_1_12_1","volume-title":"Specvlm: Fast speculative decoding in vision-language models. arXiv preprint arXiv:2509.11815","author":"Huang H.","year":"2025","unstructured":"Huang, H., Yang, F., Liu, Z., Yin, X., Li, D., Ren, P., and Barsoum, E. Specvlm: Fast speculative decoding in vision-language models. arXiv preprint arXiv:2509.11815 (2025)."},{"key":"e_1_3_2_1_13_1","volume-title":"Specvlm: Fast speculative decoding in vision-language models. arXiv preprint arXiv:2509.11815","author":"Huang H.","year":"2025","unstructured":"Huang, H., Yang, F., Liu, Z., Yin, X., Li, D., Ren, P., and Barsoum, E. Specvlm: Fast speculative decoding in vision-language models. arXiv preprint arXiv:2509.11815 (2025)."},{"key":"e_1_3_2_1_14_1","volume-title":"Know me, respond to me: Benchmarking llms for dynamic user profiling and personalized responses at scale. arXiv preprint arXiv:2504.14225","author":"Jiang B.","year":"2025","unstructured":"Jiang, B., Hao, Z., Cho, Y.-M., Li, B., Yuan, Y., Chen, S., Ungar, L., Taylor, C. J., and Roth, D. Know me, respond to me: Benchmarking llms for dynamic user profiling and personalized responses at scale. arXiv preprint arXiv:2504.14225 (2025)."},{"key":"e_1_3_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICWS67624.2025.00046"},{"key":"e_1_3_2_1_16_1","volume-title":"Vispec: Accelerating visionlanguage models with vision-aware speculative decoding. arXiv preprint arXiv:2509.15235","author":"Kang J.","year":"2025","unstructured":"Kang, J., Shu, H., Li, W., Zhai, Y., and Chen, X. Vispec: Accelerating visionlanguage models with vision-aware speculative decoding. arXiv preprint arXiv:2509.15235 (2025)."},{"key":"e_1_3_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/3672608.3707792"},{"key":"e_1_3_2_1_18_1","volume-title":"Seed-bench: Benchmarking multimodal llms with generative comprehension","author":"Li B.","year":"2023","unstructured":"Li, B.,Wang, R.,Wang, G., Ge, Y., Ge, Y., and Shan, Y. Seed-bench: Benchmarking multimodal llms with generative comprehension, 2023."},{"key":"e_1_3_2_1_19_1","volume-title":"Proceedings of the 40th International Conference on Machine Learning (2023), ICML'23, JMLR.org.","author":"Li J.","unstructured":"Li, J., Li, D., Savarese, S., and Hoi, S. Blip-2: bootstrapping language-image pretraining with frozen image encoders and large language models. In Proceedings of the 40th International Conference on Machine Learning (2023), ICML'23, JMLR.org."},{"key":"e_1_3_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/PerComWorkshops65533.2025.00078"},{"key":"e_1_3_2_1_21_1","volume-title":"Tracking meets lora: Faster training, larger model, stronger performance","author":"Lin L.","year":"2024","unstructured":"Lin, L., Fan, H., Zhang, Z., Wang, Y., Xu, Y., and Ling, H. Tracking meets lora: Faster training, larger model, stronger performance, 2024."},{"key":"e_1_3_2_1_22_1","volume-title":"Visual instruction tuning","author":"Liu H.","year":"2023","unstructured":"Liu, H., Li, C., Wu, Q., and Lee, Y. J. Visual instruction tuning, 2023."},{"key":"e_1_3_2_1_23_1","volume-title":"Aligned vector quantization for edge-cloud collabrative vision-language models. arXiv preprint arXiv:2411.05961","author":"Liu X.","year":"2024","unstructured":"Liu, X., Zhang, L., Ganesan, D., and Guan, H. Aligned vector quantization for edge-cloud collabrative vision-language models. arXiv preprint arXiv:2411.05961 (2024)."},{"key":"e_1_3_2_1_24_1","volume-title":"Mmbench: Is your multi-modal model an all-around player?","author":"Liu Y.","year":"2024","unstructured":"Liu, Y., Duan, H., Zhang, Y., Li, B., Zhang, S., Zhao, W., Yuan, Y., Wang, J., He, C., Liu, Z., Chen, K., and Lin, D. Mmbench: Is your multi-modal model an all-around player?, 2024."},{"key":"e_1_3_2_1_25_1","volume-title":"Dora: Enhancing parameter-efficient fine-tuning with dynamic rank distribution. arXiv preprint arXiv:2405.17357","author":"Mao Y.","year":"2024","unstructured":"Mao, Y., Huang, K., Guan, C., Bao, G., Mo, F., and Xu, J. Dora: Enhancing parameter-efficient fine-tuning with dynamic rank distribution. arXiv preprint arXiv:2405.17357 (2024)."},{"key":"e_1_3_2_1_26_1","volume-title":"Smolvlm: Redefining small and efficient multimodal models. arXiv preprint arXiv:2504.05299","author":"Marafioti A.","year":"2025","unstructured":"Marafioti, A., Zohar, O., Farr\u00e9, M., Noyan, M., Bakouch, E., Cuenca, P., Zakka, C., Allal, L. B., Lozhkov, A., Tazi, N., et al. Smolvlm: Redefining small and efficient multimodal models. arXiv preprint arXiv:2504.05299 (2025)."},{"key":"e_1_3_2_1_27_1","volume-title":"Eve: Efficient multimodal vision language models with elastic visual experts","author":"Rang M.","year":"2025","unstructured":"Rang, M., Bi, Z., Liu, C., Tang, Y., Han, K., and Wang, Y. Eve: Efficient multimodal vision language models with elastic visual experts, 2025."},{"key":"e_1_3_2_1_28_1","volume-title":"Share & Trends Forecast","author":"Research Nester","year":"2035","unstructured":"Research Nester. Multimodal AI Market Size, Share & Trends Forecast 2035. [Online]. Available: https:\/\/www.researchnester.com\/reports\/multimodalai-market\/6472, 2025. Accessed: 2025--10-07."},{"key":"e_1_3_2_1_29_1","volume-title":"Vision-language models for edge networks: A comprehensive survey","author":"Sharshar A.","year":"2025","unstructured":"Sharshar, A., Khan, L. U., Ullah,W., and Guizani, M. Vision-language models for edge networks: A comprehensive survey. IEEE Internet of Things Journal (2025)."},{"key":"e_1_3_2_1_30_1","volume-title":"Hydralora: An asymmetric lora architecture for efficient fine-tuning","author":"Tian C.","year":"2024","unstructured":"Tian, C., Shi, Z., Guo, Z., Li, L., and Xu, C. Hydralora: An asymmetric lora architecture for efficient fine-tuning, 2024."},{"key":"e_1_3_2_1_31_1","volume-title":"A short proof of optimality for the min cache replacement algorithm. Information processing letters 102, 2--3","author":"Van Roy B.","year":"2007","unstructured":"Van Roy, B. A short proof of optimality for the min cache replacement algorithm. Information processing letters 102, 2--3 (2007), 72--73."},{"key":"e_1_3_2_1_32_1","volume-title":"Fastvlm: Efficient vision encoding for vision language models","author":"Vasu P. K. A.","year":"2025","unstructured":"Vasu, P. K. A., Faghri, F., Li, C.-L., Koc, C., True, N., Antony, A., Santhanam, G., Gabriel, J., Grasch, P., Tuzel, O., and Pouransari, H. Fastvlm: Efficient vision encoding for vision language models, 2025."},{"key":"e_1_3_2_1_33_1","volume-title":"Vision as lora","author":"Wang H.","year":"2025","unstructured":"Wang, H., Ye, Y., Li, B., Nie, Y., Lu, J., Tang, J., Wang, Y., and Huang, C. Vision as lora, 2025."},{"key":"e_1_3_2_1_34_1","volume-title":"Spec-vla: speculative decoding for vision-language-action models with relaxed acceptance. arXiv preprint arXiv:2507.22424","author":"Wang S.","year":"2025","unstructured":"Wang, S., Yu, R., Yuan, Z., Yu, C., Gao, F., Wang, Y., and Wong, D. F. Spec-vla: speculative decoding for vision-language-action models with relaxed acceptance. arXiv preprint arXiv:2507.22424 (2025)."},{"key":"e_1_3_2_1_35_1","volume-title":"Efficientvlm: Fast and accurate vision-language models via knowledge distillation and modal-adaptive pruning","author":"Wang T.","year":"2022","unstructured":"Wang, T., Zhou, W., Zeng, Y., and Zhang, X. Efficientvlm: Fast and accurate vision-language models via knowledge distillation and modal-adaptive pruning, 2022."},{"key":"e_1_3_2_1_36_1","volume-title":"On-device language models: A comprehensive review. arXiv preprint arXiv:2409.00088","author":"Xu J.","year":"2024","unstructured":"Xu, J., Li, Z., Chen, W., Wang, Q., Gao, X., Cai, Q., and Ling, Z. On-device language models: A comprehensive review. arXiv preprint arXiv:2409.00088 (2024)."},{"key":"e_1_3_2_1_37_1","volume-title":"Visionthink: Smart and efficient vision language model via reinforcement learning","author":"Yang S.","year":"2025","unstructured":"Yang, S., Li, J., Lai, X., Yu, B., Zhao, H., and Jia, J. Visionthink: Smart and efficient vision language model via reinforcement learning, 2025."},{"key":"e_1_3_2_1_38_1","volume-title":"Train small, infer large: Memory-efficient lora training for large language models","author":"Zhang J.","year":"2025","unstructured":"Zhang, J., Wang, J., Li, H., Shou, L., Chen, K., You, Y., Xie, G., Gong, X., and Zhou, K. Train small, infer large: Memory-efficient lora training for large language models, 2025."},{"key":"e_1_3_2_1_39_1","volume-title":"Adalora: Adaptive budget allocation for parameterefficient fine-tuning","author":"Zhang Q.","year":"2023","unstructured":"Zhang, Q., Chen, M., Bukharin, A., Karampatziakis, N., He, P., Cheng, Y., Chen, W., and Zhao, T. Adalora: Adaptive budget allocation for parameterefficient fine-tuning, 2023."},{"key":"e_1_3_2_1_40_1","volume-title":"Mme-realworld: Could your multimodal llm challenge high-resolution real-world scenarios that are difficult for humans? arXiv preprint arXiv:2408.13257","author":"Zhang Y.-F.","year":"2024","unstructured":"Zhang, Y.-F., Zhang, H., Tian, H., Fu, C., Zhang, S., Wu, J., Li, F., Wang, K., Wen, Q., Zhang, Z., et al. Mme-realworld: Could your multimodal llm challenge high-resolution real-world scenarios that are difficult for humans? arXiv preprint arXiv:2408.13257 (2024)."},{"key":"e_1_3_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2858384"},{"key":"e_1_3_2_1_42_1","volume-title":"Internvl3: Exploring advanced training and test-time recipes for open-source multimodal models","author":"Zhu J.","year":"2025","unstructured":"Zhu, J., Wang, W., Chen, Z., Liu, Z., Ye, S., Gu, L., Tian, H., Duan, Y., Su, W., Shao, J., Gao, Z., Cui, E., Wang, X., Cao, Y., Liu, Y., Wei, X., Zhang, H., Wang, H., Xu, W., Li, H., Wang, J., Deng, N., Li, S., He, Y., Jiang, T., Luo, J., Wang, Y., He, C., Shi, B., Zhang, X., Shao,W., He, J., Xiong, Y., Qu,W., Sun, P., Jiao, P., Lv, H., Wu, L., Zhang, K., Deng, H., Ge, J., Chen, K., Wang, L., Dou, M., Lu, L., Zhu, X., Lu, T., Lin, D., Qiao, Y., Dai, J., and Wang, W. Internvl3: Exploring advanced training and test-time recipes for open-source multimodal models, 2025."},{"key":"e_1_3_2_1_43_1","first-page":"02","volume":"13","author":"Zhuang Z.","year":"2025","unstructured":"Zhuang, Z., Chen, J., Xu, W., Qi, Q., Guo, S., Wang, J., Lu, L., Yang, H., and Liao, J. DECC: Delay-Aware Edge-Cloud Collaboration for Accelerating DNN Inference. IEEE Transactions on Emerging Topics in Computing 13, 02 (Apr. 2025), 438--450.","journal-title":"DECC: Delay-Aware Edge-Cloud Collaboration for Accelerating DNN Inference. IEEE Transactions on Emerging Topics in Computing"}],"event":{"name":"WWW '26: The ACM Web Conference 2026","location":"Dubai United Arab Emirates","sponsor":["SIGWEB ACM Special Interest Group on Hypertext, Hypermedia, and Web"]},"container-title":["Proceedings of the ACM Web Conference 2026"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3774904.3792127","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T07:26:27Z","timestamp":1783149987000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3774904.3792127"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,12]]},"references-count":43,"alternative-id":["10.1145\/3774904.3792127","10.1145\/3774904"],"URL":"https:\/\/doi.org\/10.1145\/3774904.3792127","relation":{},"subject":[],"published":{"date-parts":[[2026,4,12]]},"assertion":[{"value":"2026-04-12","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}