cs.CVNov 14, 2026

Generate, Align, and Restore: MLLM-Empowered Diffusion for Generalizable Low-light Image Enhancement

Authors: Kaisheng Pang, Feng Li, Chen Zhang, Junjun Jiang, Haoang Li, Sam Kwong, Runmin Cong

Organizations: Thrust of Robotics and Autonomous Systems, The Hong Kong University of Science and Technology (Guangzhou), Guangzhou 511453, China · Key Laboratory of Knowledge Engineering with Big Data, Ministry of Education of China, Hefei 230601, China · Innovation School of Artificial Intelligence, Hefei University of Technology, Hefei 230601, China · School of Control Science and Engineering, Shandong University, Jinan 250061, China · 

Abstract

Low-light image enhancement (LLIE) is essential to numerous vision applications operating under adverse illumination. Despite substantial advances, real-world nighttime enhancement remains challenging due to the scarcity of strictly aligned day-night pairs and complex degradations involving extreme darkness and heterogeneous artificial light sources, leading to fragile generalization. To overcome these challenges, we propose MrLIE (MLLM-empowered reference Low-light Image Enhancement), a novel reference-based unsupervised LLIE framework that synergizes the generative and reasoning capabilities of multimodal large language models (MLLMs) with conditional diffusion. MrLIE employs the MLLM as a high-fidelity reference synthesizer and diagnostic critic, iteratively generating, diagnosing, and correcting normal-light references directly from degraded nighttime observations. Based on this, our method governs the enhancement process through two complementary mechanisms, in which a confidence offset module rectifies spatial-geometric inconsistencies for reference reliability and a structure-anchored injection scheme harmonizes aligned reference cues, low-light structural constraints, and ControlNet additive anchoring during the denoising process. Extensive experiments on diverse benchmarks demonstrate that MrLIE achieves superior perceptual quality, natural appearance, and robust generalization.

Figures & tables

Explore similar work

CardsList
  1. Internally Referenced Low-Light Enhancement

    May 27, 2026Peiyuan He, Hainuo Wang, Hengxing Liu +2Self-Supervised LearningLow-Light Image Enhancement

  2. ControlLight: Towards Controllable, Consistent, and Generalizable Low-Light Enhancement

    May 25, 2026Yufeng Yang, Jianzhuang Liu, Jisheng Chu +4Image EnhancementLow-Light Image Enhancement