Agentic Chain-of-Thought Steering for Efficient and Controllable LLM Reasoning

• URL: http://arxiv.org/abs/2606.03965v1
• Date: Tue, 02 Jun 2026 17:51:30 GMT
• Title: Agentic Chain-of-Thought Steering for Efficient and Controllable LLM Reasoning
• Abstract summary: Large language models improve final-answer accuracy through extended chain-of-thought reasoning.<n>Existing efficient reasoning methods control thinking length by shortening, early-stopping, or compressing traces.<n>We propose Agentic Chain-of-Thought Steering (ACTS), which formulates reasoning steering as a Markov decision process.
• Score: 27.851326399978394
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models improve final-answer accuracy through extended chain-of-thought reasoning, but often spend tokens inefficiently and offer little inference-time control. Existing efficient reasoning methods control thinking length by shortening, early-stopping, or compressing traces, leaving how the model thinks implicit. In this paper, we propose Agentic Chain-of-Thought Steering (ACTS), which formulates reasoning steering as a Markov decision process where a controller agent adaptively steers a frozen reasoner during inference. At each step, the controller observes the reasoning trace and remaining thinking budget, then issues a steering action consisting of a reasoning strategy and a steering phrase that initiates the next reasoner step. This enables budget-aware strategy control for efficient reasoning while preserving the reasoner's generation continuity. We initialize the controller agent from our constructed synthetic steering trajectories with multi-budget augmentation, and further optimize it via reinforcement learning with budget-conditioned reward shaping. Experiments across multiple benchmarks show that ACTS matches full-thinking performance with substantial token savings, and enables controllable accuracy-efficiency trade-offs across different reasoners and tasks. The code is available at https://github.com/Andree-9/ACTS.

Related papers

• Cognitive Demand Steering for Adaptive Meta-Reasoning in Large Language Models [12.235165912578095]
  Cognitive Demand Steering is a training-free meta-reasoning framework equipped with residual demand assessment.<n>Averaged across three frontier LLMs and six reasoning benchmarks, CDS improves accuracy by $21.9%$ over direct calls and $9%$ over standard CoT reasoning.
  arXiv  Detail & Related papers  (2026-08-02T15:41:12Z)
• Can We Break LLMs Out of Self-Loops? Fine-Grained Reasoning Control with Activation Steering [44.273459587532535]
  We analyze and discover latent transition dynamics in the reasoning traces from Large Language Models.<n>We show that failure trajectories get stuck in self-loops, exhausting the token budget without progress toward the final answer.<n>We propose SOPHIA: Steering Of reasoning Processes via Hidden-state Intervention and Activations.
  arXiv  Detail & Related papers  (2026-07-20T16:01:57Z)
• Dual-Track CoT: Budget-Aware Stepwise Guidance for Small LMs [0.3823356975862005]
  Small Language Models struggle with multi-step reasoning under tight compute and token budgets.<n>Existing test time reasoning methods such as self consistency improve performance, but often at high token cost and without fine-grained step-level control.<n>This project aims to address that gap: can Small Language Models (SLMs) reason reliably using the same or fewer tokens?
  arXiv  Detail & Related papers  (2026-04-27T22:43:33Z)
• Think When Needed: Model-Aware Reasoning Routing for LLM-based Ranking [25.69863022367215]
  Reasoning prompting can enhance ranking utility, but its benefits are inconsistent and come at a substantial computational cost.<n>We propose a reasoning routing framework that employs a lightweight, plug-and-play router head to decide whether to use direct inference (Non-Think) or reasoning (Think) for each instance before generation.
  arXiv  Detail & Related papers  (2026-01-26T05:09:07Z)
• Structured Reasoning for Large Language Models [59.215789462977206]
  We propose Structured Reasoning (SCR), a framework that decouples reasoning trajectories into explicit, evaluable, and trainable components.<n>SCR substantially improves reasoning efficiency and self-verification.<n>Compared with existing reasoning paradigms, it reduces output token length by up to 50%.
  arXiv  Detail & Related papers  (2026-01-12T04:04:01Z)
• Reinforced Efficient Reasoning via Semantically Diverse Exploration [73.41112984160992]
  Reinforcement learning with verifiable rewards (RLVR) has proven effective in enhancing the reasoning of large language models (LLMs)<n>We propose reinforced efficient reasoning via semantically diverse explorations, i.e., ROSE, for LLMs.<n>Our method incorporates a semantic-entropy-based branching strategy and an $varepsilon$-exploration mechanism.
  arXiv  Detail & Related papers  (2026-01-08T15:56:44Z)
• Fast Thinking for Large Language Models [67.7238685892317]
  We introduce Latent Codebooks for Fast Thinking, a framework that uses concise CoT sketches only during training to learn a codebook of discrete strategy priors.<n>At inference, the model conditions on a handful of continuous thinking switches distilled from the codebook in a single pass, enabling strategy-level guidance without producing explicit reasoning tokens.
  arXiv  Detail & Related papers  (2025-09-28T04:19:48Z)
• SSPO: Self-traced Step-wise Preference Optimization for Process Supervision and Reasoning Compression [15.87106741558898]
  Post-training methods incur substantial computational overhead due to auxiliary models and overthinking.<n>We propose Self-traced Step-wise Preference Optimization (SSPO), a plug RLgable process supervision framework.<n>SSPO uses step-wise preference signals generated by the model itself to guide the optimization process for reasoning compression.
  arXiv  Detail & Related papers  (2025-08-18T04:02:15Z)
• From "Aha Moments" to Controllable Thinking: Toward Meta-Cognitive Reasoning in Large Reasoning Models via Decoupled Reasoning and Control [11.321315058502215]
  Large Reasoning Models (LRMs) have demonstrated a latent capacity for complex reasoning by spontaneously exhibiting cognitive behaviors such as step-by-step reasoning, reflection, and backtracking, commonly referred to as "Aha Moments"<n>However, such emergent behaviors remain unregulated and uncontrolled, often resulting in overthinking, where the model continues generating redundant reasoning content even after reaching reliable conclusions.<n>Current models are unable to monitor and adaptively manage their reasoning process to determine when to continue, backtrack, or terminate.<n>We propose the Meta-cognitive Reasoning Framework (MERA), which explicitly decouples the thinking process into distinct
  arXiv  Detail & Related papers  (2025-08-06T13:59:17Z)
• KV Cache Steering for Controlling Frozen LLMs [80.50365534625438]
  cache steering is a lightweight method for implicit steering of language models.<n>We apply cache steering to induce chain-of-thought reasoning in small language models.
  arXiv  Detail & Related papers  (2025-07-11T17:59:36Z)
• On Reasoning Strength Planning in Large Reasoning Models [50.61816666920207]
  We find evidence that LRMs pre-plan the reasoning strengths in their activations even before generation.<n>We then uncover that LRMs encode this reasoning strength through a pre-allocated directional vector embedded in the activations of the model.<n>Our work provides new insights into the internal mechanisms of reasoning in LRMs and offers practical tools for controlling their reasoning behaviors.
  arXiv  Detail & Related papers  (2025-06-10T02:55:13Z)
• Adaptive Deep Reasoning: Triggering Deep Thinking When Needed [28.575411507835973]
  Large language models (LLMs) have shown impressive capabilities in handling complex tasks through long-chain reasoning.<n>We propose a novel approach that autonomously switches between short and long-chain reasoning chains based on problem complexity.<n>This advancement enhances the practicality of reasoning in large language models for real-world applications.
  arXiv  Detail & Related papers  (2025-05-26T15:08:51Z)
• Let LRMs Break Free from Overthinking via Self-Braking Tuning [68.93713497579853]
  Large reasoning models (LRMs) have significantly enhanced their reasoning capabilities by generating longer chains of thought.<n>This performance gain comes at the cost of a substantial increase in redundant reasoning during the generation process.<n>We propose a novel framework, Self-Braking Tuning (SBT), which tackles overthinking from the perspective of allowing the model to regulate its own reasoning process.
  arXiv  Detail & Related papers  (2025-05-20T16:53:40Z)

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.