{"ID":2850148,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.22838","arxiv_id":"2510.22838","title":"Semantic-Preserving Cross-Style Visual Reasoning for Robust Multi-Modal Understanding in Large Vision-Language Models","abstract":"The \"style trap\" poses a significant challenge for Large Vision-Language Models (LVLMs), hindering robust semantic understanding across diverse visual styles, especially in in-context learning (ICL). Existing methods often fail to effectively decouple style from content, hindering generalization. To address this, we propose the Semantic-Preserving Cross-Style Visual Reasoner (SP-CSVR), a novel framework for stable semantic understanding and adaptive cross-style visual reasoning. SP-CSVR integrates a Cross-Style Feature Encoder (CSFE) for style-content disentanglement, a Semantic-Aligned In-Context Decoder (SAICD) for efficient few-shot style adaptation, and an Adaptive Semantic Consistency Module (ASCM) employing multi-task contrastive learning to enforce cross-style semantic invariance. Extensive experiments on a challenging multi-style dataset demonstrate SP-CSVR's state-of-the-art performance across visual captioning, visual question answering, and in-context style adaptation. Comprehensive evaluations, including ablation studies and generalization analysis, confirm SP-CSVR's efficacy in enhancing robustness, generalization, and efficiency across diverse visual styles.","short_abstract":"The \"style trap\" poses a significant challenge for Large Vision-Language Models (LVLMs), hindering robust semantic understanding across diverse visual styles, especially in in-context learning (ICL). Existing methods often fail to effectively decouple style from content, hindering generalization. To address this, we pr...","url_abs":"https://arxiv.org/abs/2510.22838","url_pdf":"https://arxiv.org/pdf/2510.22838v1","authors":"[\"Aya Nakayama\",\"Brian Wong\",\"Yuji Nishimura\",\"Kaito Tanaka\"]","published":"2025-10-26T21:11:46Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Language Model\"]","has_code":false}