Looped Flows: diffusion-style denoising plus recurrent reasoning hits 58.8% on ARC-AGI-1
LucaAmb · x · 2026-09-14
A new arXiv paper, Thinking with Looped Flows, proposes Looped Flows, combining flow/diffusion-style denoising with recurrent reasoning. It trains the recurrence with local denoising objectives, sidestepping long BPTT, and uses progressively decreasing noise levels with shared noise to incentivize hidden states that transfer useful computation over time. Inference integrates the velocity of a probability flow parameterized by the learned denoiser, so harder problems can be solved with a finer temporal grid, and multiple valid predictions can be drawn from different initial noise samples. Across six reasoning benchmarks (including two multi-solution ones), looped flows beat prior state-of-the-art looped models overall, reaching 58.8% on ARC-AGI-1 and 12.2% on ARC-AGI-2.
Related event: Looped Flows Paper Combines Diffusion Denoising with Recurrent Reasoning(2 posts)→
More from Research
- Swapping pretraining objective cuts entity-swap false-accepts from 46% to 5% with zero training — Reasonable_Royal_621 · 2026-09-14
- LeanDB: Theoric Labs builds a strongly typed Lean frontend for SQL databases — hargup13 · 2026-09-14
- DeepMind looks back on 15 years of AI game research, partners with EVE Online devs — arnicas · 2026-09-14
- ECCV paper demystifies video reasoning: diffusion steps hide parallel trajectories — ziqi_huang_ · 2026-09-14
- Should arXiv reject an AI-discovered cancer breakthrough that passes clinical trials? — IanArawjo · 2026-09-14
- cESA: judging multiple translations at once cuts human eval time and noise — zouharvi · 2026-09-14