FineSteer: A Unified Framework for Fine-Grained Inference-Time Steering in Large Language Models

TL;DR

FineSteer introduces a two-stage inference-time steering framework for large language models, enhancing safety and truthfulness while preserving utility and training efficiency.

Contribution

It proposes a novel, flexible steering method with conditional steering and vector synthesis, outperforming existing approaches in effectiveness and efficiency.

Findings

Outperforms state-of-the-art methods on safety and truthfulness benchmarks.

Maintains high utility with minimal performance loss.

Demonstrates adaptive, training-efficient steering for targeted inputs.

Abstract

Large language models (LLMs) often exhibit undesirable behaviors, such as safety violations and hallucinations. Although inference-time steering offers a cost-effective way to adjust model behavior without updating its parameters, existing methods often fail to be simultaneously effective, utility-preserving, and training-efficient due to their rigid, one-size-fits-all designs and limited adaptability. In this work, we present FineSteer, a novel steering framework that decomposes inference-time steering into two complementary stages: conditional steering and fine-grained vector synthesis, allowing fine-grained control over when and how to steer internal representations. In the first stage, we introduce a Subspace-guided Conditional Steering (SCS) mechanism that preserves model utility by avoiding unnecessary steering. In the second stage, we propose a Mixture-of-Steering-Experts (MoSE) mechanism that captures the multimodal nature of desired steering behaviors and generates query-specific steering vectors for improved effectiveness. Through tailored designs in both SCS and MoSE, FineSteer maintains robust performance on general queries while adaptively optimizing steering vectors for targeted inputs in a training-efficient manner. Extensive experiments on safety and truthfulness benchmarks show that FineSteer outperforms state-of-the-art methods in overall performance, achieving stronger steering performance with minimal utility loss. Code is available at https://github.com/YukinoAsuna/FineSteer