{"type":"ScholarlyArticle","title":"DSSG: Dual-Stream Semantic Guidance for Source-Fully-Free Adaptation of Vision-Language Models","authors":[{"name":"Weiwei Xiang","slug":"weiwei-xiang","given":"Weiwei","family":"Xiang","affiliation":""},{"name":"Shun Peng","slug":"shun-peng","given":"Shun","family":"Peng","affiliation":""},{"name":"Guangyi Xiao","slug":"guangyi-xiao","given":"Guangyi","family":"Xiao","affiliation":""},{"name":"Hao Chen","slug":"hao-chen","given":"Hao","family":"Chen","affiliation":""},{"name":"Lei Yang","slug":"lei-yang","given":"Lei","family":"Yang","affiliation":""}],"abstract":"Source-Fully-Free Domain Adaptation (SFF-DA) has emerged as a strategic paradigm to adapt Vision-Language Models (VLMs) without any access to source data or task-specific source models. However, we identify a critical \"Dual Semantic Drift\" that hinders this process: static drift caused by the stagnation of frozen hand-crafted templates, and dynamic drift resulting from noisy, instance-level generated captions. These issues lead to a severe stability-plasticity dilemma, manifesting as semantic misalignment and class collapse.    \r\nTo address this, we propose DSSG (Dual-Stream Semantic Guidance), an end-to-end framework that reconciles fine-grained plasticity with global stability. Our core contribution is the Dual Semantic Guidance (DSG) module, which integrates a captioning stream for detailed domain nuances with a template stream to anchor global categorical consistency. Furthermore, a Dynamic Cross-Modal Knowledge Distillation (CMKD) module is introduced to adaptively optimize teacher-student alignment across modalities. Extensive experiments demonstrate that DSSG significantly outperforms current state-of-the-art methods across multiple benchmarks, providing a robust solution for the unsupervised transfer of foundation models. The code is available on https://github.com/mrmenand/DSSG.","doi":"10.24963/ijcai.2026/211","url":"https://ijcai.org/proceedings/2026/211","pdf":"https://ijcai.org/proceedings/2026/0211.pdf","markdown":"https://ijcai.org/proceedings/2026/211.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/211","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Computer Vision","track":"Main Track","number":211,"pages":{"first":1893,"last":1901},"keywords":["Computer Vision: Transfer, low-shot, semi- and un- supervised learning","Computer Vision: Multimodal learning"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}