Know-Thy-Not: Training-Free Targeted Negation in Dual-Encoder Vision-Language Models

arXiv:2610.17440 · cs.CV, cs.AI · Submitted 2026-10-12 · Read on arXiv

cs.CV, cs.AI

Submitted: 2026-10-12

Updated: 2026-10-12

Comments: 9 pages

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Vision-language models — 1× in this paper · explained in 2 episodes

Sources

• Hierarchical Text-Conditional Image Generation with CLIP Latents
• Learn "No" to Say "Yes" Better: Improving Vision-Language Models via Negations
• SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models