Effectively Steer LLM To Follow Preference via Building Confident Directions

• URL: http://arxiv.org/abs/2503.02989v1
• Date: Tue, 04 Mar 2025 20:32:27 GMT
• Title: Effectively Steer LLM To Follow Preference via Building Confident Directions
• Authors: Bingqing Song, Boran Han, Shuai Zhang, Hao Wang, Haoyang Fang, Bonan Min, Yuyang Wang, Mingyi Hong,
• Abstract summary: We propose a theoretical framework to understand and quantify the model steering methods.<n>Inspired by the framework, we propose a confident direction steering method (CONFST) that steers LLMs via modifying their activations.<n>Our approach offers three key advantages over popular bidirectional model steering methods.
• Score: 39.40603123075168
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Having an LLM that aligns with human preferences is essential for accommodating individual needs, such as maintaining writing style or generating specific topics of interest. The majority of current alignment methods rely on fine-tuning or prompting, which can be either costly or difficult to control. Model steering algorithms, which modify the model output by constructing specific steering directions, are typically easy to implement and optimization-free. However, their capabilities are typically limited to steering the model into one of the two directions (i.e., bidirectional steering), and there has been no theoretical understanding to guarantee their performance. In this work, we propose a theoretical framework to understand and quantify the model steering methods. Inspired by the framework, we propose a confident direction steering method (CONFST) that steers LLMs via modifying their activations at inference time. More specifically, CONFST builds a confident direction that is closely aligned with users' preferences, and this direction is then added to the activations of the LLMs to effectively steer the model output. Our approach offers three key advantages over popular bidirectional model steering methods: 1) It is more powerful, since multiple (i.e. more than two) users' preferences can be aligned simultaneously; 2) It is simple to implement, since there is no need to determine which layer to add the steering vector to; 3) No explicit user instruction is required. We validate our method on GPT-2 XL (1.5B), Mistral (7B) and Gemma-it (9B) models for tasks that require shifting the output of LLMs across various topics and styles, achieving superior performance over competing methods.

Related papers

• Self-Steering Language Models [113.96916935955842]
  DisCIPL is a method for "self-steering" language models. DisCIPL uses a Planner model to generate a task-specific inference program. Our work opens up a design space of highly-parallelized Monte Carlo inference strategies.
  arXiv  Detail & Related papers  (2025-04-09T17:54:22Z)
• Steering Large Language Models with Feature Guided Activation Additions [0.0]
  We introduce Feature Guided Activation Additions (FGAA), a novel activation steering method.<n>By operating in the latent space of a Sparse Autoencoder (SAE), FGAA constructs precise steering vectors.<n> evaluations on Gemma-2-2B and Gemma-2-9B models demonstrate that FGAA outperforms existing steering methods.
  arXiv  Detail & Related papers  (2025-01-17T02:55:23Z)
• Improving Instruction-Following in Language Models through Activation Steering [58.876600545898675]
  We derive instruction-specific vector representations from language models and use them to steer models accordingly. We demonstrate how this method can enhance model adherence to constraints such as output format, length, and word inclusion. Our findings demonstrate that activation steering offers a practical and scalable approach for fine-grained control in language generation.
  arXiv  Detail & Related papers  (2024-10-15T08:38:20Z)
• Aligning Large Language Models with Representation Editing: A Control Perspective [38.71496554018039]
  Fine-tuning large language models (LLMs) to align with human objectives is crucial for real-world applications. Test-time alignment techniques, such as prompting and guided decoding, do not modify the underlying model. We propose aligning LLMs through representation editing.
  arXiv  Detail & Related papers  (2024-06-10T01:21:31Z)
• Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization [34.05163996072159]
  "steering vectors" are extracted from the activations of human preference data. This work proposes an innovative approach that could produce more effective steering vectors through bi-directional preference optimization. Our method is designed to allow steering vectors to directly influence the generation probability of contrastive human preference data pairs.
  arXiv  Detail & Related papers  (2024-05-28T05:10:40Z)
• InferAligner: Inference-Time Alignment for Harmlessness through Cross-Model Guidance [56.184255657175335]
  We develop textbfInferAligner, a novel inference-time alignment method that utilizes cross-model guidance for harmlessness alignment. Experimental results show that our method can be very effectively applied to domain-specific models in finance, medicine, and mathematics. It significantly diminishes the Attack Success Rate (ASR) of both harmful instructions and jailbreak attacks, while maintaining almost unchanged performance in downstream tasks.
  arXiv  Detail & Related papers  (2024-01-20T10:41:03Z)
• LanguageMPC: Large Language Models as Decision Makers for Autonomous Driving [87.1164964709168]
  This work employs Large Language Models (LLMs) as a decision-making component for complex autonomous driving scenarios. Extensive experiments demonstrate that our proposed method not only consistently surpasses baseline approaches in single-vehicle tasks, but also helps handle complex driving behaviors even multi-vehicle coordination.
  arXiv  Detail & Related papers  (2023-10-04T17:59:49Z)
• Parameter-Efficient Tuning Helps Language Model Alignment [57.27390187540737]
  Previous works mainly adopt reinforcement learning (RLHF) and direct preference optimization (DPO) with human feedback for alignment. Controllable generation offers more flexibility with regard to data format. Our approach, alignMEnt with parameter-Efficient Tuning (MEET), improves the quality of control tokens.
  arXiv  Detail & Related papers  (2023-10-01T23:27:14Z)
• Guiding Large Language Models via Directional Stimulus Prompting [114.84930073977672]
  We introduce Directional Stimulus Prompting, a novel framework for guiding black-box large language models (LLMs) toward specific desired outputs. Instead of directly adjusting LLMs, our method employs a small tunable policy model to generate an auxiliary directional stimulus prompt for each input instance.
  arXiv  Detail & Related papers  (2023-02-22T17:44:15Z)

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.