Vision-Language Models (VLMs) struggle with negation. Given a prompt like "retrieve (or generate) a street scene without pedestrians," they often fail to respect the "not." Existing methods address this limitation by fine-tuning on large negation datasets, but such retraining often compromises the model's zero-shot performance on affirmative prompts. We show that the embedding space of VLMs, such as CLIP, can be divided into semantically consistent subspaces. Based on this property, we propose a training-free framework that models negation as a subspace in the joint embedding space rather than a single point (Figure 1). To find the matching image for a caption such as "A but not N," we construct two spherical caps around the embeddings of A and N, and we score images by the central direction of the region that is close to A and far from N. Across retrieval, MCQ, and text-to-image tasks, our method improves negation understanding by about 30% on average over prior methods. It closes the gap between affirmative and negated prompts while preserving the zero-shot performance that fine-tuned models fail to maintain. Code will be released upon publication.
Negation-Aware Test-Time Adaptation for Vision-Language Models
Department of AI Computing, Pengcheng Laboratory（人工智能计算系，鹏城实验室）；School of Computer Science and Technology, Central South University（计算机科学与技术学院，中南大学）；Department of Computer Science and Technology, Institute for AI, BNRist Center, THBI Lab, Tsinghua-Bosch Joint Center for ML, Tsinghua University（计算机科学与技术系，人工智能研究院，BNRist中心，THBI实验室，清华-博世联合人工智能中心，清华大学）
HANCLIP: A Family of Hyperbolic Angular Negation Vision Language Models
ADAPT Centre Dublin City University, Ireland（爱尔兰都柏林城市大学ADAPT中心）；University of East Anglia Norwich, UK（英国东英吉利大学）；Queen’s University Belfast Belfast, UK（英国贝尔法斯特女王大学）；University of Science Vietnam National University Ho Chi Minh City, Vietnam（越南胡志明市国家大学理科大学）
Bi-MCQ: Reformulating Vision-Language Alignment for Negation Understanding
Department of Electrical and Computer Engineering, Inha University, Republic of Korea（电气与计算机工程系，印哈大学，大韩民国）；College of Medicine, Inha University, Republic of Korea（医学学院，印哈大学，大韩民国）
University of Arkansas（阿肯色大学）；University of Science, VNU-HCM（胡志明市国家大学科学大学）；Aalto University（阿尔托大学）；SpexAI GmbH（SpexAI有限公司）；University of California, San Diego（加利福尼亚大学圣迭戈分校）
Scale Can't Overcome Pragmatics: The Impact of Reporting Bias on Vision-Language Reasoning
University of Washington（华盛顿大学）；University of California, Los Angeles（加州大学洛杉矶分校）；Samaya AI（Samaya人工智能）；Mistral AI（Mistral人工智能）；Allen Institute for AI（人工智能研究所）
EchoVLM: Measurement-Grounded Multimodal Learning for Echocardiography
Georgia Institute of Technology（佐治亚理工学院）；Siemens Healthineers（西门子医疗）；Siemens Healthcare Limited（西门子医疗有限公司）；Emory University（埃默里大学）；University of Southern California（南加州大学）
Bringing CLIP to the Clinic: Dynamic Soft Labels and Negation-Aware Learning for Medical Analysis
Interdisciplinary Program in Bioengineering, Seoul National University Graduate School（生物工程跨学科项目，首尔国立大学研究生院）；Integrated Major in Innovative Medical Science, Seoul National University Graduate School（创新医学联合专业，首尔国立大学研究生院）；Department of Radiology, Seoul National University Hospital（放射科，首尔国立大学医院）
