{"ID":2894856,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.10225","arxiv_id":"2507.10225","title":"Synthesizing Near-Boundary OOD Samples for Out-of-Distribution Detection","abstract":"Pre-trained vision-language models have exhibited remarkable abilities in detecting out-of-distribution (OOD) samples. However, some challenging OOD samples, which lie close to in-distribution (InD) data in image feature space, can still lead to misclassification. The emergence of foundation models like diffusion models and multimodal large language models (MLLMs) offers a potential solution to this issue. In this work, we propose SynOOD, a novel approach that harnesses foundation models to generate synthetic, challenging OOD data for fine-tuning CLIP models, thereby enhancing boundary-level discrimination between InD and OOD samples. Our method uses an iterative in-painting process guided by contextual prompts from MLLMs to produce nuanced, boundary-aligned OOD samples. These samples are refined through noise adjustments based on gradients from OOD scores like the energy score, effectively sampling from the InD/OOD boundary. With these carefully synthesized images, we fine-tune the CLIP image encoder and negative label features derived from the text encoder to strengthen connections between near-boundary OOD samples and a set of negative labels. Finally, SynOOD achieves state-of-the-art performance on the large-scale ImageNet benchmark, with minimal increases in parameters and runtime. Our approach significantly surpasses existing methods, and the code is available at https://github.com/Jarvisgivemeasuit/SynOOD.","short_abstract":"Pre-trained vision-language models have exhibited remarkable abilities in detecting out-of-distribution (OOD) samples. However, some challenging OOD samples, which lie close to in-distribution (InD) data in image feature space, can still lead to misclassification. The emergence of foundation models like diffusion model...","url_abs":"https://arxiv.org/abs/2507.10225","url_pdf":"https://arxiv.org/pdf/2507.10225v3","authors":"[\"Jinglun Li\",\"Kaixun Jiang\",\"Zhaoyu Chen\",\"Bo Lin\",\"Yao Tang\",\"Weifeng Ge\",\"Wenqiang Zhang\"]","published":"2025-07-14T12:43:50Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\",\"Large Language Model\",\"Language Model\"]","has_code":false,"code_links":[{"ID":612139,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2894856,"paper_url":"https://arxiv.org/abs/2507.10225","paper_title":"Synthesizing Near-Boundary OOD Samples for Out-of-Distribution Detection","repo_url":"https://github.com/Jarvisgivemeasuit/SynOOD","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}