Loading
Life sciences · Preprint
arXiv · September 10, 2026
Posted before peer review. The findings may change or fail to hold.
This is an unrefereed preprint describing looped flows, a machine learning method that trains recurrent models using local denoising objectives to solve reasoning tasks. The work reports benchmark results on six reasoning tasks but has not undergone peer review and is not applicable to clinical practice or medical decision-making.
Preprint. Intervention: Looped flows: a method training recurrent models through local denoising objectives with temporal association across progressively decreasing noise levels and shared noise, with inference via velocity integration of a probability flow.. Compared with: Prior state-of-the-art looped models.
Looped flows achieved 58.8% test accuracy on ARC-AGI-1 Looped flows achieved 12.2% test accuracy on ARC-AGI-2 Method outperforms prior state-of-the-art looped models overall across six reasoning benchmarks
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
The source did not state who this applies to in practice.
This is an unrefereed arXiv preprint presenting a novel machine learning method with benchmark results; it lacks peer review and clinical or regulatory validation.
As stated by the source record.
Quoted from the source exactly as published.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
Humans and machines often solve harder problems by spending more time on computation. In deep learning, looped models implement this idea during inference by recurrently updating a hidden state. In practice, however, their training backpropagates through only one or a few updates, making it hard to train early updates to support future ones. We propose looped flows, an approach that sidesteps this issue by training the recurrence with local denoising objectives. By imposing temporal association across denoising objectives through progressively decreasing noise levels and shared noise, the model is incentivized to learn recurrent states that transfer useful computation over time, even when gradients cover only a few updates. We then formulate inference as integrating the velocity of a probability flow parameterized by the learned denoiser, coupled with recurrent states. This allows solving harder problems by spending more computation through a finer temporal grid and enables multiple valid predictions from different initial noise samples. Across six reasoning benchmarks including two multi-solution benchmarks, looped flows outperform prior state-of-the-art looped models overall, achieving 58.8% test accuracy on ARC-AGI-1 and 12.2% on ARC-AGI-2.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.