AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Bilevel Layer-Positioning LoRA for Real Image Dehazing

Yan Zhang, Long Ma, Yuxin Feng, Zhe Huang, Fan Zhou, Zhuo Su

CVPR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Unimodal-to-multimodal representation transfer · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add positive text guidance to H2C loss (no negative, no positioning) (ablation)DEA adaptation · Real dehazing averageFADE: 1.018 -> 0.862Improved
add negative text guidance alone (ablation)DEA adaptation · Real dehazing averageMUSIQ: 62.05 -> 60.47Hurt
add both positive and negative text guidance versus positive only (ablation)DEA adaptation · Real dehazing averageMUSIQ: 62.23 -> 63.31Improved
add naive (joint) layer positioning to full H2C loss (ablation)DEA adaptation · Real dehazing averageFADE: 0.774 -> 0.662Improved
replace naive joint positioning with bilevel positioning under full H2C (ablation)DEA adaptation · Real dehazing averageFADE: 0.662 -> 0.638Improved
replace naive joint positioning with bilevel positioning under full H2C (ablation)DEA adaptation · Real dehazing averageMUSIQ: 64.07 -> 64.40Improved
add bilevel positioning with positive-only H2C (ablation)DEA adaptation · Real dehazing averageBIQME: 0.589 -> 0.601Improved
replace full fine-tuning with BiLaLoRADEA adaptation · Full fine-tuning vs BiLaLoRATraining time (hours): 4.215 -> 0.940Improved
replace full fine-tuning with BiLaLoRADEA adaptation · Full fine-tuning vs BiLaLoRAFADE: 0.610 -> 0.638Hurt
replace full fine-tuning with BiLaLoRADEA adaptation · Full fine-tuning vs BiLaLoRAMUSIQ: 64.43 -> 64.40No clear effect
increase LoRA layers up to three (ablation)BiLaLoRA · Real dehazing averageOverall qualityImproved
apply BiLaLoRA to synthetic-trained DEADEA · Real dehazing averageMUSIQ: 57.42 -> 64.40Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy