Dual Flow: joint image-measurement flow model hits SOTA on linear inverse problems
Sevket Ege Yesilyurt · hf · 2026-11-17
A new HF paper introduces Dual Flow, a joint-flow model for linear inverse problems that co-estimates the clean image and noise-free measurement via coupled flows, guaranteeing forward-model consistency by construction. The authors derive closed-form sufficient statistics showing the needed conditional mean reduces to a Gaussian denoising problem with operator-dependent correlated noise. Experiments show state-of-the-art distortion and perceptual quality with no hyperparameter tuning beyond inference steps and faster reconstruction than competing methods.
More from Research
- BrickFlow uses SE(3) flow models to reconstruct buildable LEGO assemblies from images — Peter Kulits · 2026-11-17
- Meta-model writes weight updates from one paragraph, beating LoRA on many benchmarks — gdm3000 · 2026-11-17
- SPARE preprint: 2.55x faster decoding for linear attention on a 1.3B model — krishnan · 2026-11-17
- Study estimates ~12% of philosophy published today may be AI-generated — leland_mcinnes · 2026-11-17
- Berkeley EECS & Statistics open all-rank faculty search in AI for Physical Sciences — berkeley_ai · 2026-11-17
- Origami as physical reservoir computing: powering robotic crawling in Scientific Reports — cephaloform · 2026-11-17