---
title: "Orthogonal and Parametrized Multi-Attribute Steering of LLMs"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.37517v1"
permalink: "https://feed.astralor.com/items/itm_57ce1c14e7bd905562da7b05"
published: "2026-10-27T16:39:53.000Z"
score: 51
channel: "tech"
tags: ["LLM","推理效率","Orthogonal","Parametrized Multi-Attribute Steering","LLMs Activation","However","Activation"]
---

# Orthogonal and Parametrized Multi-Attribute Steering of LLMs

## At a glance

Activation steering offers an effective and efficient way to modify an LLM's behavior at inference time. However, it remains unclear how to apply steering such that the signal is consistently strong and the steering str…

## Key points

- Activation steering offers an effective and efficient way to modify an LLM's behavior at inference time.
- However, it remains unclear how to apply steering such that the signal is consistently strong and the steering strength has a universally interpretable meaning.
- We address this first by identifying the most steerable interval for each steering vector and rescaling the steering range accordingly.

## Full text

Activation steering offers an effective and efficient way to modify an LLM's behavior at inference time. However, it remains unclear how to apply steering such that the signal is consistently strong and the steering strength has a universally interpretable meaning. We address this first by identifying the most steerable interval for each steering vector and rescaling the steering range accordingly. Building on this, we introduce a more interpretable method for steering, called parameterized steering which instead of pushing activations in a direction, it sets them at the desired trait strength. Next, with multi-trait steering, we find that naively combining vectors causes them to interfere with one another. In response, we develop a method to orthogonalize the vectors; it decouples these traits, allowing a model to exhibit one behavior without inducing the others, thus enabling multiple traits to be steered simultaneously.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_57ce1c14e7bd905562da7b05