titleTailSteer: Trajectory-Tail Steering for Drift-Resistant Multi-turn Image Editing
authorsRuikang Li, Zixuan Chen, Jiarui Wu, Mingde Yao, Yiyuan Zhang, Yujin Wang, Hanming Deng, Lewei Lu, Tianfan Xue
date2026-11-04
venuearXiv (preprint, not peer reviewed)
idarXiv:2611.06741
affiliationMMLab-CUHK
upvotes39 (Hugging Face Daily Papers, 2026-11-08)
linksabs pdf hf code
cite
abstractStrong single-turn editing performance does not guarantee reliable multi-turn interaction: as edited outputs are repeatedly reused, small unintended changes can accumulate into severe visual drift. This long-term drift is often mitigated through training on multi-turn data or cross-turn rollouts, potentially altering pretrained priors, or through specialized inference-time mechanisms that are difficult to scale to foundation models. We introduce TailSteer, a final-stage training framework that improves multi-turn robustness while retaining the generative and editing capabilities acquired through earlier training. TailSteer follows a simple principle: keep early denoising dynamics close to those of the base model and concentrate preservation corrections near the trajectory tail, where fine-grained deviations can be suppressed. These corrections are learned from the editor's own on-policy denoising rollouts, using unchanged source content as a preservation reference without requiring curated ground-truth editing pairs. Experiments across the FLUX, SenseNova, and Qwen families show that TailSteer substantially reduces cumulative drift over 15-20 editing turns while largely preserving base-model performance on extensive text-to-image and single-image editing benchmarks. These results demonstrate that long-horizon editing robustness can be unlocked in foundation models after TailSteer, rather than by relearning their existing capabilities.

bibtex

@misc{li2026tailsteer,
  title={TailSteer: Trajectory-Tail Steering for Drift-Resistant Multi-turn Image Editing},
  author={Ruikang Li and Zixuan Chen and Jiarui Wu and Mingde Yao and Yiyuan Zhang and Yujin Wang and Hanming Deng and Lewei Lu and Tianfan Xue},
  year={2026},
  eprint={2611.06741},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2611.06741}
}