Controllable Reasoning Models Are Private Thinkers

TL;DR

This paper introduces a method to improve privacy in reasoning models by training them to follow instructions in reasoning traces, leading to significant privacy gains but with some utility trade-offs.

Contribution

The paper proposes a novel training approach that enhances privacy preservation in reasoning models by controlling their reasoning traces through instruction-following.

Findings

Up to 20.9 point improvement in instruction-following performance.

Up to 51.9 percentage points improvement on privacy benchmarks.

Decoupling reasoning and answer generation enhances privacy.

Abstract

AI agents powered by reasoning models require access to sensitive user data. However, their reasoning traces are difficult to control, which can result in the unintended leakage of private information to external parties. We propose training models to follow instructions not only in the final answer, but also in reasoning traces, potentially under different constraints. We hypothesize that improving their instruction following abilities in the reasoning traces can improve their privacy-preservation skills. To demonstrate this, we fine-tune models on a new instruction-following dataset with explicit restrictions on reasoning traces. We further introduce a generation strategy that decouples reasoning and answer generation using separate LoRA adapters. We evaluate our approach on six models from two model families, ranging from 1.7B to 14B parameters, across two instruction-following benchmarks and two privacy benchmarks. Our method yields substantial improvements, achieving gains of up to 20.9 points in instruction-following performance and up to 51.9 percentage points on privacy benchmarks. These improvements, however, can come at the cost of task utility, due to the trade-off between reasoning performance and instruction-following abilities. Overall, our results show that improving instruction-following behavior in reasoning models can significantly enhance privacy, suggesting a promising direction for the development of future privacy-aware agents. Our code and data are available at https://github.com/UKPLab/arxiv2026-controllable-reasoning-models