FIRM embeds the measurement operator in a flow's velocity field, cutting NFEs up to 50x
prof_kamilov · x · 2026-10-07
Ulugbek Kamilov's group (WashU) introduces FIRM (Flow-based Imaging via Regularized Minimization), a new approach to imaging inverse problems that embeds the forward measurement operator directly into the conditional velocity field. The key insight: the optimal conditional velocity is determined by a posterior mean, which they characterize as the unique minimizer of a variational objective with an explicit data-consistency term. The resulting operator-aware velocity field is trained end-to-end and requires no separate guidance at sampling time.
Related event: FIRM Cuts Network Evaluations 50x for Inverse Imaging(3 posts)→
More from Research
- Hidden Chain-of-Thought Extracted from GPT-6 and Claude via Standard API Tool Calling — jiqizhixin · 2026-10-07
- Information-Dense Synthesis cuts molecular discovery experiments from O(d) to O(log d) — anshulkundaje · 2026-10-07
- UCLA finds hybrid-attention layer ordering shapes multilinguality — alternative orders learn 2.5X faster — UCLA · 2026-10-07
- NYU's RGPO uses adaptive rationale scaffolding to fight reward sparsity in RL for reasoning — newyorkuniversity · 2026-10-07
- UCLA aligns cross-lingual representations in decoder-only LLMs via MoE router outputs — UCLA · 2026-10-07
- 12M+ RL task entries on HF Hub, browsable via new RL Environment Explorer — SergioPaniego · 2026-10-07