Hugging Face Daily Papers·· 2026-10-15AI score42
ACEdit: stable multi-turn image editing with asymmetric conditioning
ACEdit: Stable Multi-Turn Image Editing via Asymmetric Conditioning
AI briefing
ACEdit proposes a training scheme that keeps image editing models stable across multiple consecutive editing turns. Its core is introducing Self-Forcing into image editing: during training the editor is forced to process its own previous output, while a teacher editor provides distillation signals; the teacher is always anchored to the original clean image and receives composite editing instructions, forming asymmetric conditioning. Experiments show significantly reduced quality degradation over long multi-turn sequences while maintaining editing quality and instruction following.
Open original arxiv.org