# A Mechanistic Analysis of Looped Reasoning Language Models

arXiv: https://arxiv.org/abs/2604.11791 · HTML: https://mlsanity.com/paper/2604.11791/

## Key facts

- Cited by 33 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 34 of its 51 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 30 of 34)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Universal Transformers](https://mlsanity.com/paper/1807.03819/) (2018)
- [On the Bottleneck of Graph Neural Networks and its Practical Implications](https://mlsanity.com/paper/2006.05205/) (2020)
- [Adaptive Computation Time for Recurrent Neural Networks](https://mlsanity.com/paper/1603.08983/) (2016)
- [Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach](https://mlsanity.com/paper/2502.05171/) (2025)
- [A Note on Over-Smoothing for Graph Neural Networks](https://mlsanity.com/paper/2006.13318/) (2020)
- [Reasoning with Latent Thoughts: On the Power of Looped Transformers](https://mlsanity.com/paper/2502.17416/) (2025)
- [When Attention Sink Emerges in Language Models: An Empirical View](https://mlsanity.com/paper/2410.10781/) (2024)
- [Hierarchical Reasoning Model](https://mlsanity.com/paper/2506.21734/) (2025)
- [Scaling Latent Reasoning via Looped Language Models](https://mlsanity.com/paper/2510.25741/) (2025)
- [PonderNet: Learning to Ponder](https://mlsanity.com/paper/2107.05407/) (2021)
- [Looped Transformers are Better at Learning Learning Algorithms](https://mlsanity.com/paper/2311.12424/) (2023)
- [The Remarkable Robustness of LLMs: Stages of Inference?](https://mlsanity.com/paper/2406.19384/) (2024)
- [Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation](https://mlsanity.com/paper/2507.10524/) (2025)
- [Why do LLMs attend to the first token?](https://mlsanity.com/paper/2504.02732/) (2025)
- [Attention Sinks and Compression Valleys in LLMs are Two Sides of the Same Coin](https://mlsanity.com/paper/2510.06477/) (2025)
- [Teaching Pretrained Language Models to Think Deeper with Retrofitted Recurrence](https://mlsanity.com/paper/2511.07384/) (2025)
- [Less is More: Recursive Reasoning with Tiny Networks](https://mlsanity.com/paper/2510.04871/) (2025)
- [Softmax is not Enough (for Sharp Size Generalisation)](https://mlsanity.com/paper/2410.01104/) (2024)
- [On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learning](https://mlsanity.com/paper/2502.10818/) (2025)
- [On Expressive Power of Looped Transformers: Theoretical Analysis and Enhancement via Timestep Encoding](https://mlsanity.com/paper/2410.01405/) (2024)
- [Sparse Universal Transformer](https://mlsanity.com/paper/2310.07096/) (2023)
- [Latent Chain-of-Thought? Decoding the Depth-Recurrent Transformer](https://mlsanity.com/paper/2507.02199/) (2025)
- [Encode, Think, Decode: Scaling test-time reasoning with recursive latent thoughts](https://mlsanity.com/paper/2510.07358/) (2025)
- [Softmax is 1/2-Lipschitz: A tight bound across all ℓp norms](https://mlsanity.com/paper/2510.23012/) (2025)
- [Continuous Thought Machines](https://mlsanity.com/paper/2505.05522/) (2025)
- [Two-Scale Latent Dynamics for Recurrent-Depth Transformers](https://mlsanity.com/paper/2509.23314/) (2025)
- [Return of ChebNet: Understanding and Improving an Overlooked GNN on Long Range Tasks](https://mlsanity.com/paper/2506.07624/) (2025)

## Built on by (most cited first, 30 of 33)

- [Fixed-Point Reasoners: Stable and Adaptive Deep Looped Transformers](https://mlsanity.com/paper/2606.18206/) (2026)
- [Equilibrium Reasoners: Learning Attractors Enables Scalable Reasoning](https://mlsanity.com/paper/2605.21488/) (2026)
- [Memory-Efficient Looped Transformer: Decoupling Compute from Memory in Looped Language Models](https://mlsanity.com/paper/2605.07721/) (2026)
- [Solve the Loop: Attractor Models for Language and Reasoning](https://mlsanity.com/paper/2605.12466/) (2026)
- [Hyperloop Transformers](https://mlsanity.com/paper/2604.21254/) (2026)
- [Thinking with Looped Flows](https://mlsanity.com/paper/2609.11801/) (2026)
- [Loop the Loopies!](https://mlsanity.com/paper/2607.16051/) (2026)
- [LT2: Linear-Time Looped Transformers](https://mlsanity.com/paper/2605.20670/) (2026)
- [Looped Latent Attention: Cross-Loop KV Compression for Looped Transformers](https://mlsanity.com/paper/2607.15456/) (2026)
- [When Does Recurrence Become an Algorithm? Convergence Selection in Weight-Tied Looped Transformers](https://mlsanity.com/paper/2607.20594/) (2026)
- [Looped Transformers under the Jacobian Lens: Does the Global Workspace Survive Recurrence?](https://mlsanity.com/paper/2609.01924/) (2026)
- [Dense Supervision Is Not Enough: The Readout Blind Spot in Looped Language Models](https://mlsanity.com/paper/2606.24898/) (2026)
- [On the Diverse Dynamical Behaviors Arising in Deep Linear Transformers](https://mlsanity.com/paper/2607.18584/) (2026)
- [Chain-of-Thought and Compressed Looped Transformers: A Memory-Budget Separation](https://mlsanity.com/paper/2605.30757/) (2026)
- [Mechanistic Analysis of Alignment Algorithms in Language Models](https://mlsanity.com/paper/2606.09850/) (2026)
- [Steering Recurrent Reasoners at Inference Time with Readout Feedback](https://mlsanity.com/paper/2608.24136/) (2026)
- [Improving Test-Time Scaling with Adaptive Looped Transformers](https://mlsanity.com/paper/2609.35748/) (2026)
- [Towards Looped Models Done Right, Part II: Rethinking at Fixed Points](https://mlsanity.com/paper/2610.06833/) (2026)
- [SMolLM: Small Language Models Learn Small Molecular Grammar](https://mlsanity.com/paper/2605.06322/) (2026)
- [Reshape and Recur: Improving SSMs with Input Reshaping and Depth Recurrence](https://mlsanity.com/paper/2605.16048/) (2026)
- [One Model, Two Roles: Emergent Specialization in a Shared Recurrent Transformer](https://mlsanity.com/paper/2605.17811/) (2026)
- [Behaviour Is an Incomplete Measure of Reasoning Development: Cross-surface pre-arrival accessibility and the limits of developmental inference in a recurrent-depth reasoner](https://mlsanity.com/paper/2608.16085/) (2026)
- [If It's Not Buggy, Don't Fix It: On the Dynamics of Iterative Bug-fixing with LLMs](https://mlsanity.com/paper/2609.10123/) (2026)
- [Beyond the Training Horizon: Mechanisms and Limits of Length Generalization in Looped Transformers](https://mlsanity.com/paper/2609.33144/) (2026)
- [LoopICL: Looping a single transformer block to solve tabular tasks](https://mlsanity.com/paper/2609.36108/) (2026)
- [Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It](https://mlsanity.com/paper/2609.36585/) (2026)
- [What Makes Recurrence Effective in Looped Language Models?](https://mlsanity.com/paper/2609.36636/) (2026)
- [LoopVL: Recurrent Visual Intelligence](https://mlsanity.com/paper/2609.38426/) (2026)
- [Shared Weights, Selected Computations: How Looped Transformers Route What Each Loop Does](https://mlsanity.com/paper/2609.39892/) (2026)
- [SanSi: A Looped Typed Decision Model for System 1.5 Thinking](https://mlsanity.com/paper/2610.07730/) (2026)

## Abstract (arXiv)

Reasoning has become a central capability in large language models. Recent research has shown that reasoning performance can be improved by looping an LLM's layers in the latent dimension, resulting in looped reasoning language models. Despite promising results, few works have investigated how their internal dynamics differ from those of standard feedforward models. In this paper, we conduct a mechanistic analysis of the latent states in looped language models, focusing in particular on how the stages of inference observed in feedforward models compare to those observed in looped ones. To this end, we analyze cyclic recurrence and show that for many of the studied models each layer in the cycle converges to a distinct fixed point; consequently, the recurrent block follows a consistent cyclic trajectory in the latent space. We provide evidence that as these fixed points are reached, attention-head behavior stabilizes, leading to constant behavior across recurrences. Empirically, we discover that recurrent blocks learn stages of inference that closely mirror those of feedforward models, repeating these stages in depth with each iteration. We study how recurrent block size, input injection, and normalization influence the emergence and stability of these cyclic fixed points. We believe these findings help translate mechanistic insights into practical guidance for architectural design.