English

Mosaic: Compositional Multi-Concept Erasure via Vector Field Blending

Computer Vision and Pattern Recognition 2026-05-26 v1 Artificial Intelligence

Abstract

Concept erasure has emerged as a key research direction for ensuring safe and ethical image synthesis in Text-to-Image (T2I) models. While existing studies have explored concept erasure across multiple concepts, they typically assume only a single target concept per image, a limitation increasingly exposed by modern flow-based T2I models, which can generate complex scenes with multiple concepts simultaneously. To address this gap, we introduce compositional multi-concept erasure, a new task that aims to simultaneously remove multiple target concepts within a single scene. We propose CoME-Bench, a benchmark for evaluating compositional multi-concept erasure, which covers both intra- and cross-category scenarios. We further propose Mosaic, a novel framework for multi-concept erasure in flow-based T2I models, which exploits the spatial locality of target concepts in the vector field by dynamically constructing concept-specific masks and selectively blending them without additional optimization. Extensive experiments demonstrate that Mosaic effectively removes multiple target concepts in complex compositional scenes while preserving non-target contexts.

Cite

@article{arxiv.2605.25574,
  title  = {Mosaic: Compositional Multi-Concept Erasure via Vector Field Blending},
  author = {Junseok Ko and Jungwoo Kim and Jong-Seok Lee},
  journal= {arXiv preprint arXiv:2605.25574},
  year   = {2026}
}
R2 v1 2026-07-22T07:32:03.323Z