Steering Recurrent Reasoners at Inference Time with Readout Feedback

• URL: http://arxiv.org/abs/2608.24136v1
• Date: Tue, 25 Aug 2026 07:00:21 GMT
• Title: Steering Recurrent Reasoners at Inference Time with Readout Feedback
• Abstract summary: We show that recurrent models can be improved at inference time by using their own readout probabilities to steer latent dynamics without retraining.<n>Across three recurrent models (AKOrN, ItrSA++, TRM) on Sudoku and Maze, RoFB yields clear gains in four of six model-task pairs.
• Score: 22.32377205563293
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recurrent models, which repeatedly update latent states with shared computation blocks, have emerged as powerful architectures for solving complex reasoning tasks. Existing inference-time methods scale computation by running more steps or sampling more trajectories, but ignore information revealed within each trajectory. Here we show that recurrent models can be improved at inference time by using their own readout probabilities to steer latent dynamics without retraining. We introduce Readout Feedback (RoFB), a test-time intervention that converts intermediate predictions into token-wise pairwise coupling forces injected into the latent dynamics. Across three recurrent models (AKOrN, ItrSA++, TRM) on Sudoku and Maze, RoFB yields clear gains in four of six model-task pairs, achieving performance unattainable by merely running more steps or selecting from multiple trajectories, at comparable or lower computational cost. These results suggest that closed-loop steering of latent dynamics can serve as a complementary inference-time control mechanism for recurrent reasoning models.

Related papers

• Recursive Think-Answer Process for LLMs and VLMs [54.52289112197118]
  We propose an efficient Recursive Think-Answer Process (R-TAP)<n>R-TAP enables models to engage in iterative reasoning cycles and generate more accurate answers.<n>We show that R-TAP-enhanced models consistently outperform conventional single-pass methods.
  arXiv  Detail & Related papers  (2026-03-02T17:20:10Z)
• Reactive Knowledge Representation and Asynchronous Reasoning [35.58961985804191]
  Exact inference in complex probabilistic models often incurs prohibitive computational costs.<n>Existing methods are often inefficient for ongoing reasoning, as they re-evaluate the entire model upon any change.<n>We first introduce Resin, a probabilistic programming language that merges probabilistic logic with reactive programming.<n>To provide efficient and exact semantics for Resin, we propose Reactive Circuits (RCs)<n>In high-fidelity drone swarm simulations, our approach achieves several orders of magnitude of speedup over frequency-agnostic inference.
  arXiv  Detail & Related papers  (2026-02-05T13:02:01Z)
• NEAT: Neuron-Based Early Exit for Large Reasoning Models [37.88270573682948]
  Large Reasoning Models (LRMs) often suffer from emphoverthinking, a phenomenon in which redundant reasoning steps are generated after a correct solution has been reached.<n>Existing early reasoning exit methods rely on output-levels or trained probing models to skip redundant reasoning steps.<n>We propose textbfNEAT, a framework that monitors neuron-level activation dynamics to enable training-free early exits.
  arXiv  Detail & Related papers  (2026-02-02T12:09:59Z)
• Drift No More? Context Equilibria in Multi-Turn LLM Interactions [58.69551510148673]
  contexts drift is the gradual divergence of a model's outputs from goal-consistent behavior across turns.<n>Unlike single-turn errors, drift unfolds temporally and is poorly captured by static evaluation metrics.<n>We show that multi-turn drift can be understood as a controllable equilibrium phenomenon rather than as inevitable decay.
  arXiv  Detail & Related papers  (2025-10-09T04:48:49Z)
• Parallel Test-Time Scaling for Latent Reasoning Models [58.428340345068214]
  Parallel test-time scaling (TTS) is a pivotal approach for enhancing large language models (LLMs)<n>Recent advances in latent reasoning, where intermediate reasoning unfolds in continuous vector spaces, offer a more efficient alternative to explicit Chain-of-Thought.<n>This work enables parallel TTS for latent reasoning models by addressing the above issues.
  arXiv  Detail & Related papers  (2025-10-09T03:33:00Z)
• Inverse Scaling in Test-Time Compute [51.16323216811257]
  Extending the reasoning length of Large Reasoning Models (LRMs) deteriorates performance.<n>We identify five distinct failure modes when models reason for longer.<n>These findings suggest that while test-time compute scaling remains promising for improving model capabilities, it may inadvertently reinforce problematic reasoning patterns.
  arXiv  Detail & Related papers  (2025-07-19T00:06:13Z)
• SPRINT: Enabling Interleaved Planning and Parallelized Execution in Reasoning Models [2.7645012830234]
  Large reasoning models excel at complex reasoning tasks but typically generate lengthy sequential chains-of-thought.<n>SPRINT is a novel post-training and inference-time framework designed to enable LRMs to dynamically identify and exploit opportunities for parallelization.<n>We show that the models fine-tuned with the SPRINT framework match the performance of reasoning models on complex domains such as mathematics.
  arXiv  Detail & Related papers  (2025-06-06T05:10:31Z)
• Gated Recurrent Neural Networks with Weighted Time-Delay Feedback [55.596897987498174]
  We present a novel approach to modeling long-term dependencies in sequential data by introducing a gated recurrent unit (GRU) with a weighted time-delay feedback mechanism.<n>Our proposed model, named $tau$-GRU, is a discretized version of a continuous-time formulation of a recurrent unit, where the dynamics are governed by delay differential equations (DDEs)
  arXiv  Detail & Related papers  (2022-12-01T02:26:34Z)
• Deep Explicit Duration Switching Models for Time Series [84.33678003781908]
  We propose a flexible model that is capable of identifying both state- and time-dependent switching dynamics. State-dependent switching is enabled by a recurrent state-to-switch connection. An explicit duration count variable is used to improve the time-dependent switching behavior.
  arXiv  Detail & Related papers  (2021-10-26T17:35:21Z)
• Causal Modeling with Stochastic Confounders [11.881081802491183]
  This work extends causal inference with confounders. We propose a new approach to variational estimation for causal inference based on a representer theorem with a random input space.
  arXiv  Detail & Related papers  (2020-04-24T00:34:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.