ControlLight: Towards Controllable, Consistent, and Generalizable Low-Light Enhancement
Yufeng Yang Jianzhuang Liu Jisheng Chu Yuqi Peng, [0.4em] Xianfang Zeng, Jiancheng Huang, Shifeng Chen, [0.4em] Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences · 2026
Read ControlLight: Towards Controllable, Consistent, and Generalizable Low-Light Enhancement by Yufeng Yang Jianzhuang Liu Jisheng Chu Yuqi Peng, [0.4em] Xianfang Zeng, Jiancheng Huang, Shifeng Chen, [0.4em] Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-25569.txt.
Description
Yufeng Yang and colleagues give users a dial for brightening dark photos instead of a single fixed result. Earlier enhancers train on narrow sets with one target, limiting control and generalization. ControlLight trains on a large set of real degraded images labeled with continuous illumination strength, and a misalignment-aware flow-matching loss keeps structure steady as the dial moves. Users can therefore tune strength freely while outputs stay consistent and realistic, and tests report state-of-the-art quality with strong generalization.
Catalog record
| CREATOR | Yufeng Yang Jianzhuang Liu Jisheng Chu Yuqi Peng, [0.4em] Xianfang Zeng, Jiancheng Huang, Shifeng Chen, [0.4em] Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35511 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.25569