From Leaky Thoughts to Private Reasoning: Controlling What LRMs Say to Themselves

arXiv:2602.24210 · cs.CL, cs.AI · Submitted 2026-02-27 · Read on arXiv

cs.CL, cs.AI

Submitted: 2026-02-27

Updated: 2026-08-28

Code: https://github.com/UKPLab/arxiv2026-controllable-reasoning-models

Terminology

• Trace — 36× in this paper · explained in 2 episodes
• Benchmark — 34× in this paper · explained in 2 episodes
• Reasoning Traces — 21× in this paper · explained in 3 episodes
• LoRA Adapters — 7× in this paper · explained in 3 episodes
• Information Leakage — 6× in this paper · explained in 2 episodes
• Accuracy — 4× in this paper · explained in 1 episode
• Alignment — 3× in this paper · explained in 1 episode
• Fine-Tuning — 3× in this paper · explained in 5 episodes
• In-Context Learning — 3× in this paper · explained in 4 episodes
• Quantization — 3× in this paper · explained in 12 episodes
• Reinforcement Learning — 3× in this paper · explained in 4 episodes
• Adversarial Attacks — 2× in this paper · explained in 1 episode
• LLM-as-a-Judge — 2× in this paper · explained in 3 episodes
• Steering — 2× in this paper · explained in 2 episodes
• Adaptation — 1× in this paper · explained in 1 episode
• Adaptive Reasoning — 1× in this paper · explained in 2 episodes
• Adaptive Selection — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 3 episodes
• Computational overhead — 1× in this paper · explained in 2 episodes
• Core — 1× in this paper · explained in 2 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• LLM judge — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Parameter-Efficient Fine-Tuning — 1× in this paper · explained in 2 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes
• Task Decomposition — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes
• alpha — 1× in this paper · explained in 1 episode

Sources

• Training Verifiers to Solve Math Word Problems
• Chain-of-Sanitized-Thoughts: Plugging PII Leakage in CoT of Large Reasoning Models
• Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning Models
• Phi-4 Technical Report
• Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation
• From "Aha Moments" to Controllable Thinking: Toward Meta-Cognitive Reasoning in Large Reasoning Models via Decoupled Reasoning and Control
• ReasonIF: Large Reasoning Models Fail to Follow Instructions During Reasoning
• Efficient Multi-Adapter LLM Serving via Cross-Model KV-Cache Reuse with Activated LoRA
• Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety
• CIMemories: A Compositional Benchmark for Contextual Integrity of Persistent Memory in LLMs
• gpt-oss-120b & gpt-oss-20b Model Card
• Effectively Controlling Reasoning Models through Thinking Intervention
• Qwen3 Technical Report
• Dynamic Early Exit in Reasoning Models
• Controlling What You Share: Assessing Language Model Adherence to Privacy Preferences
• Trade-offs in Large Reasoning Models: An Empirical Analysis of Deliberative and Adaptive Reasoning over Foundational Capabilities
• Evaluating Language Model Reasoning about Confidential Information
• Instruction-Following Evaluation for Large Language Models

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering