Thinking with Looped Flows hits SoTA on ARC-AGI-1/2 among looped models
LucaAmb · x · 2026-09-12
Researchers unveil Thinking with Looped Flows, a new reasoning approach that trains recurrent reasoning with local denoising objectives — letting neural networks "think" via iterative denoising, extending the flow-based language model agenda.
- The method achieves SoTA on both ARC-AGI-1 and ARC-AGI-2 among looped models.
- Work by @ayhozade with collaborators including David Lee and FEijkelboom.
Related event: Looped Flows: Denoising-Trained Recurrent Reasoning Tops Loop-Model ARC-AGI(6 posts)→
More from Research
- Just Action Model: decoupling what separates VLA from world-action models — Yonghao Yu · 2026-10-28
- HABIT-Bench: Best agent memory system scores just 33.61% at inferring user habits — SJTU · 2026-10-28
- CodeCap-Bench: executable code beats natural language as video captions — Jiafu Tang · 2026-10-28
- Omni-InteractScope benchmarks proactive interaction in continuous omnimodal streams — Qwen · 2026-10-28
- Mistral releases Timestral time-series foundation models, tops GIFT-Eval — mistralai · 2026-10-28
- NVIDIA's ThinkWorld learns robot skills from unlabeled human videos via chain-of-dynamics reasoning — nvidia · 2026-10-28