The paper proposes InfiLoop, a loop-native residual connection designed for looped Transformers to prevent performance degradation as iterations increase. InfiLoop learns which past computations to retain and how much to accept from each new update, using content-based weighting and learned temporal decay to maintain a running summary of recurrent states. The approach yields strong results on reasoning tasks, achieving 97.9% exact accuracy on Sudoku-Extreme and 13.6% pass@2 on ARC-AGI-2, and continues to improve with over 20,000 test-time iterations.
The paper introduces InfiLoop, a loop-native residual connection designed for looped Transformers to prevent performance degradation as iterations increase. InfiLoop learns which past computations to retain and how much to accept from each new update, combining content-based weighting with learned temporal decay to maintain a running summary of recurrent states. Across extensive reasoning tasks, a 7M-parameter InfiLoop model outperforms existing recursive architectures, achieving 97.9% exact accuracy on Sudoku-Extreme and 13.6% pass@2 on ARC-AGI-2, and continues to improve with test-time looping beyond 20,000 effective steps.
By Pengxiang Li, Dilxat Muhtar, Di He, Guinan Su, Lu Yin, Shiwei Liu
arXiv:2606. 18206v1 Announce Type: new Abstract: Looped architectures provide an inductive bias toward learning step-by-step procedures for tasks that require compositional reasoning.
By Sajad Movahedi, Vera Milovanovi\'c, Shlomo Libo Feigin, Alexander Theus, Thomas Hofmann, Valentina Boeva, T. Konstantin Rusch, Antonio Orvieto
The paper introduces LOOM, a method for scaling looped mixture‑of‑experts (MoE) Transformers beyond the typical two‑loop limit. LOOM addresses two key obstacles: it stabilizes deep recurrence by bounding residual variance and re‑injecting the input embedding, and it prevents expert selection collapse by using per‑loop routers and a looping residual to maintain computational diversity. Experiments on 100 M–1.7 B parameter models show stable scaling to 9–12 loops, with significant perplexity reductions and zero‑shot accuracy gains under near‑iso‑FLOP conditions.
By Di He, Pengxiang Li, Da Chang, Qingyan Meng, Lu Yin, Shiwei Liu
arXiv:2605. 09165v2 Announce Type: replace Abstract: Looped language models repeat a set of transformer layers through depth, reducing memory costs and providing natural early-exit points at loop boundaries.
By Ryan Lee, Jacob Biloki, Edward J. Hu, Jonathan May
arXiv:2609.36636v1 Announce Type: new
Abstract: Looped language models (LoopLMs) increase computational depth through parameter sharing, offering a path to scale inference computation without adding...
By Xinlin Zhuang, Siyuan Wang, Imran Razzak, Weiyang Liu
arXiv:2609. 39892v1 Announce Type: new Abstract: Looped Transformers repeatedly apply the same set of Transformer layers, giving them a recurrent architecture for latent computation.
By Jiaju Wu, Yi Hu, Muhan Zhang
Looped transformers and Mixture-of-Experts (MoE) offer complementary routes to efficient scaling: recurrence increases computational depth at fixed parameters, while MoE sparsity expands total capacit...
arXiv:2609.37379v1 Announce Type: new
Abstract: Looped Transformers provide a parameter-efficient approach to depth scaling by repeatedly applying shared Transformer blocks. Recent reasoning models h...
By Yulong Huang, Chen Jiang, Zhanpeng Zhou, Hongtao Zhang, Tianyu Li, Tianyu He, Xiangyu Zhang, Bojun Cheng
arXiv:2606. 12364v1 Announce Type: new Abstract: Transformers dominate modern sequence modeling, but their quadratic attention incurs substantial computational cost.
By Anamaria-Roberta Hartl, Levente Z\'olyomi, David Stap, Pieter-Jan Hoedt, Niklas Schmidinger, Lukas Hauzenberger, Sebastian B\"ock, G\"unter Klambauer, Sepp Hochreiter
arXiv:2609.40316v1 Announce Type: cross
Abstract: Looped transformers and Mixture-of-Experts (MoE) offer complementary routes to efficient scaling: recurrence increases computational depth at fixed p...
By Yanbei Chen, Anirudh Goyal, Raghuraman Krishnamoorthi
The paper introduces recirculation, an inference‑time architectural enhancement for foundation models that reduces perplexity and improves accuracy on generation and reasoning tasks without adding significant latency. Recirculation adds a specific form of recurrence, enabling the model to function as a dynamical system that tracks belief states, and is distinct from chain‑of‑thought or depth‑recurrence methods. An adaptive variant requires minimal hyperparameter tuning and achieves notable gains on the Gemma3 family, including a 23% perplexity drop and a 21% accuracy increase on GSM8k.
By Michael C. Mozer, Shoaib Ahmed Siddiqui, Danny Sawyer, Sunny Sanyal, Rosanne Liu