Looped Flows: New Training Method Boosts Looped-Model Reasoning on 5 of 6 Benchmarks
omarsar0 · x · 2026-09-13
A new paper introduces Looped Flows, a looped architecture with a novel training method: local denoising objectives borrowed from flow models, with stepwise-decreasing shared noise samples that tie each hidden-state update to the next. This fixes the usual problem that gradients only reach the last updates, and the authors report wins over prior looped models on five of six reasoning benchmarks—more reasoning without extra parameters.
Related event: Looped Flows: denoising-based recurrent reasoning tops ARC-AGI(3 posts)→
More from Research
- Same LLM eval, different scores: a method to tell real regressions from noise — bgoncalves · 2026-09-13
- "Why do AIs hack? Because we taught them to": new piece traces reward hacking to training — mmitchell_ai · 2026-09-13
- Got 2x A100 for 3 Months: How to Distill Offline Edge Models for Field Work — Lerok-Persea · 2026-09-13
- Researcher cautions third-party AI audit ecosystem is harder than it looks, cites mature audit regimes — mmitchell_ai · 2026-09-13
- TRACES opens problem and AI-solver submissions, publishes technical report — rohanpaul_ai · 2026-09-13
- Harness, not model, moves coding agent from 30th to top 5 on Terminal-Bench — blaizedsouza · 2026-09-13