LoopUS: Recasting Pretrained LLMs into Looped Latent Refinement Models

arXiv:2605.11011v1 Announce Type: cross Abstract: Looped computation shows promise in improving the reasoning-oriented performance of LLMs by scaling test-time compute. However, existing approaches typically require either training recurrent models from scratch or applying disruptive retrofits, which involve substantial computational costs and may compromise pretrained capabilities. To address these limitations, we introduce \textbf{Looped Depth Up-Scaling} (LoopUS), a post-training framework th