Reliable Control-Point Selection for Steering Reasoning in Large Language Models

• URL: http://arxiv.org/abs/2604.02113v1
• Date: Thu, 02 Apr 2026 14:48:56 GMT
• Title: Reliable Control-Point Selection for Steering Reasoning in Large Language Models
• Abstract summary: Steering vectors offer a training-free mechanism for controlling reasoning behaviors in large language models.<n>However, constructing effective vectors requires identifying genuine behavioral signals in the model's hidden states.<n>We show that current methods detect these behaviors through keyword matching in chain-of-thought traces, implicitly assuming that every detected boundary encodes a genuine behavioral signal.<n>We develop a probabilistic model that formalizes intrinsic reasoning behaviors as events with context-dependent trigger probabilities, and show that unstable boundaries dilute the steering signal.
• Score: 28.288321095634128
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Steering vectors offer a training-free mechanism for controlling reasoning behaviors in large language models, but constructing effective vectors requires identifying genuine behavioral signals in the model's hidden states. For behaviors that can be toggled via prompts, this is straightforward. However, many reasoning behaviors -- such as self-reflection -- emerge spontaneously and resist prompt-level control. Current methods detect these behaviors through keyword matching in chain-of-thought traces, implicitly assuming that every detected boundary encodes a genuine behavioral signal. We show that this assumption is overwhelmingly wrong: across 541 keyword-detected boundaries, 93.3\% are behaviorally unstable, failing to reproduce the detected behavior under re-generation from the same prefix. We develop a probabilistic model that formalizes intrinsic reasoning behaviors as stochastic events with context-dependent trigger probabilities, and show that unstable boundaries dilute the steering signal. Guided by this analysis, we propose stability filtering, which retains only boundaries where the model consistently reproduces the target behavior. Combined with a content-subspace projection that removes residual question-specific noise, our method achieves 0.784 accuracy on MATH-500 (+5.0 over the strongest baseline). The resulting steering vectors transfer across models in the same architecture family without re-extraction, improving Nemotron-Research-Reasoning-1.5B (+5.0) and DeepScaleR-1.5B-Preview (+6.0). Code is available at https://github.com/zhmzm/stability-steering.

Related papers

• Minimizing Targeted Activations: Input-Only Suppression of Evaluation-Awareness Latents in Large Language Models [0.4783156204372467]
  Activation steering controls model behavior by editing internal activations at inference time.<n>We study its input-side dual: optimizing a fluent prompt so that a chosen internal latent is driven toward zero.
  arXiv  Detail & Related papers  (2026-07-28T16:01:48Z)
• SPARK: Susceptibility-Guided Profiling and Steering of Latent Reasoning States in Large Language Models [26.849417177804302]
  We introduce SPARK, which uses hidden-state response to diagnose whether a model enters an effective reasoning state.<n>Steering can serve not only as a diagnostic signal for reasoning failures, but also as a practical guide for targeted test-time intervention.
  arXiv  Detail & Related papers  (2026-07-11T12:47:27Z)
• Internal-State Probes Read the Situation, Not the Action: Three Negative Results for Pre-Action Misalignment Monitoring [1.3636639880574484]
  Internal readouts could help monitor agentic systems if they identify harmful text or tool actions before those actions are generated.<n>We test three methods across three model families: a Qwen2.5-Coder-32B-Instruct fine-tune/base direction, Llama-3.1-8B-Instruct probes at the last token of unsafe prefills, and Gemma-3-27B-IT emotion-concept vectors used for projection and steering.
  arXiv  Detail & Related papers  (2026-06-29T15:18:36Z)
• Detecting and Controlling Sycophancy with Cascading Linear Features [7.13923185919936]
  We present an iterative data generation pipeline that isolates cascading linear features responsible for a behavior.<n>We focus on detecting and steering away from sycophancy -- the tendency of language models to prioritize user validation.
  arXiv  Detail & Related papers  (2026-06-23T20:10:53Z)
• CIAware-Bench: Benchmarking Control Intervention Awareness Across Frontier LLMs [100.38986535324284]
  We introduce textbfCIAware-Bench, a benchmark for measuring textbfcontrol textbfintervention (CI) awareness across frontier models.<n>CIAware-Bench tests whether models can distinguish their own trajectories from those modified by a control intervention.
  arXiv  Detail & Related papers  (2026-06-09T16:24:16Z)
• CaliDist: Calibrating Large Language Models via Behavioral Robustness to Distraction [51.56484100374058]
  Existing calibration methods for Large Language Models (LLMs) often overlook a critical dimension of trustworthiness: a model's behavioral robustness.<n>We introduce textscCaliDist, a novel post-hoc calibration approach that directly measures and penalizes a model's susceptibility to distraction.<n>textscCaliDist consistently achieves lower Expected Error (ECE) and Brier Score compared with strong baselines.
  arXiv  Detail & Related papers  (2026-06-04T07:27:53Z)
• Behavioral Steering in a 35B MoE Language Model via SAE-Decoded Probe Vectors: One Agency Axis, Not Five Traits [0.0]
  We train nine sparse autoencoders (SAEs) on the residual stream of Qwen 3.5-35B-A3B.<n>We use them to identify and steer five agentic behavioral traits.
  arXiv  Detail & Related papers  (2026-03-17T10:05:41Z)
• Internalizing LLM Reasoning via Discovery and Replay of Latent Actions [4.830503861275364]
  Internalization of chain-of-thought processes into hidden states has emerged as a highly efficient paradigm for scaling test-time compute.<n>We propose STIR (Self-Distilled Tools for Internal Reasoning), a framework that reformulates reasoning enhancement as a dynamic latent trajectory control problem.
  arXiv  Detail & Related papers  (2026-02-04T08:44:57Z)
• The Hypocrisy Gap: Quantifying Divergence Between Internal Belief and Chain-of-Thought Explanation via Sparse Autoencoders [0.0]
  We introduce the Hypocrisy Gap to quantify the divergence between a model's internal reasoning and its final generation.<n>By mathematically comparing an internal truth belief, derived via sparse linear probes, to the final generated trajectory in latent space, we quantify and detect a model's tendency to engage in unfaithful behavior.
  arXiv  Detail & Related papers  (2026-01-14T00:40:40Z)
• How Does Prefix Matter in Reasoning Model Tuning? [57.69882799751655]
  We fine-tune three R1 series models across three core model capabilities: reasoning (mathematics), coding, safety, and factuality.<n>Results show that prefix-conditioned SFT improves both safety and reasoning performance, yielding up to +6% higher Safe@1 accuracy.
  arXiv  Detail & Related papers  (2026-01-04T18:04:23Z)
• The 'Sure' Trap: Multi-Scale Poisoning Analysis of Stealthy Compliance-Only Backdoors in Fine-Tuned Large Language Models [10.377264470934843]
  Backdoor attacks on large language models typically couple a secret trigger to an explicit malicious output.<n>We introduce a compliance-only backdoor: supervised fine-tuning on a mostly benign dataset in which a small subset of prompts is suffixed with an arbitrary single-word trigger.<n>We conduct a multi-scale analysis of this benign-label poisoning behavior across poison budget, total fine-tuning dataset size, and model size.
  arXiv  Detail & Related papers  (2025-11-16T02:01:58Z)
• SEAL: Steerable Reasoning Calibration of Large Language Models for Free [58.931194824519935]
  Large Language Models (LLMs) have demonstrated compelling capabilities for complex reasoning tasks via the extended chain-of-thought (CoT) reasoning mechanism.<n>Recent studies reveal substantial redundancy in the CoT reasoning traces, which negatively impacts model performance.<n>We introduce SEAL, a training-free approach that seamlessly calibrates the CoT process, improving accuracy while demonstrating significant efficiency gains.
  arXiv  Detail & Related papers  (2025-04-07T02:42:07Z)
• Steering Without Side Effects: Improving Post-Deployment Control of Language Models [61.99293520621248]
  Language models (LMs) have been shown to behave unexpectedly post-deployment. We present KL-then-steer (KTS), a technique that decreases the side effects of steering while retaining its benefits. Our best method prevents 44% of jailbreak attacks compared to the original Llama-2-chat-7B model.
  arXiv  Detail & Related papers  (2024-06-21T01:37:39Z)
• Lazy Layers to Make Fine-Tuned Diffusion Models More Traceable [70.77600345240867]
  A novel arbitrary-in-arbitrary-out (AIAO) strategy makes watermarks resilient to fine-tuning-based removal. Unlike the existing methods of designing a backdoor for the input/output space of diffusion models, in our method, we propose to embed the backdoor into the feature space of sampled subpaths. Our empirical studies on the MS-COCO, AFHQ, LSUN, CUB-200, and DreamBooth datasets confirm the robustness of AIAO.
  arXiv  Detail & Related papers  (2024-05-01T12:03:39Z)

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.