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

Yuan Tian, Ying Li, Jinghuai Zhang, Peiran Wang, Zixuan Weng, Kunlin Cai

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 1 functions out of this paper's own repositories and ran 0 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

Repositories linked to this paper

• YukinoAsuna/FineSteer

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

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, utilitypreserving, 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 finegrained 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 the state-of-the-art methods in overall performance (e.g., A 7.6% improvement on TruthfulQA over Llama-3.), achieving stronger steering performance with minimal utility loss. The code is available at https://github.com/YukinoAsuna/FineSteer.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2604.15488")
get_code_for_paper("2604.15488")
have("2604.15488")

Connect an agent — have() is free.