Meta^n presents a recursive self-improvement approach for LLM agents that applies a fixed meta-operation repeatedly to its own outputs, enabling deeper reasoning layers without destabilization. The system outperforms prior self-improving agents across benchmarks, notably achieving the only non-zero score on ARC-AGI-2, with gains primarily from inter-layer conditioning.
Looped flows is a deep learning approach that trains recurrent models using local denoising objectives to enable better multi-step reasoning during inference. By coupling denoising with recurrent state updates and probability flow integration, the method achieves state-of-the-art performance on reasoning benchmarks, including 58.8% accuracy on ARC-AGI-1.