UniSteer: Text-Guided Flow Matching in Activation Space for Versatile LLM Steering

Hugging Face Daily Papers Papers
llm-steering activation-engineering flow-matching text-guided large-language-models conditional-generation activation-space

Summary

UniSteer introduces a text-guided activation flow matching method to learn a universal conditional velocity field in activation space, enabling versatile LLM behavior control and classification tasks without task-specific intervention modules.

Activation-based control steers large language models (LLMs) by intervening on their internal representations during inference, and has emerged as an effective paradigm for controlling behaviors such as persona and style. However, existing methods often rely on fixed steering directions or task-specific intervention modules, making them difficult to adapt to fine-grained concepts and compositional constraints. We propose UniSteer, a text-guided activation flow matching model that learns a conditional distribution over residual-stream activations from natural-language conditions. Instead of fitting a separate intervention for each target behavior, UniSteer learns a universal conditional velocity field in activation space. At inference time, UniSteer performs flow inversion by partially transporting a source activation toward a latent state and regenerating it under a target textual condition before injecting it back into the frozen LLM. The same conditional model supports activation-space classification by selecting the textual label with the lowest reconstruction energy. Experiments on three target LLMs show that UniSteer provides a unified interface across behavioral control, truthfulness steering, fine-grained concept steering, multi-constraint instruction following, and activation-space classification.
Original Article
View Cached Full Text

Cached at: 05/29/26, 11:01 AM

Paper page - UniSteer: Text-Guided Flow Matching in Activation Space for Versatile LLM Steering

Source: https://huggingface.co/papers/2605.30076

Abstract

UniSteer uses text-guided activation flow matching to learn a universal conditional velocity field in activation space for controlling LLM behaviors and classification tasks.

Activation-based controlsteerslarge language models(LLMs) by intervening on their internal representations during inference, and has emerged as an effective paradigm for controlling behaviors such as persona and style. However, existing methods often rely on fixed steering directions or task-specific intervention modules, making them difficult to adapt to fine-grained concepts and compositional constraints. We propose UniSteer, atext-guided activation flow matchingmodel that learns aconditional distributionoverresidual-stream activationsfrom natural-language conditions. Instead of fitting a separate intervention for each target behavior, UniSteer learns auniversal conditional velocity fieldin activation space. At inference time, UniSteer performsflow inversionby partially transporting a source activation toward a latent state and regenerating it under a target textual condition before injecting it back into the frozen LLM. The same conditional model supportsactivation-space classificationby selecting the textual label with the lowest reconstruction energy. Experiments on three target LLMs show that UniSteer provides a unified interface across behavioral control,truthfulness steering,fine-grained concept steering,multi-constraint instruction following, andactivation-space classification.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2605\.30076

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.30076 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.30076 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.30076 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

When is Your LLM Steerable?

arXiv cs.CL

This paper investigates when activation steering succeeds or fails for LLMs by analyzing early decoding dynamics. The authors introduce ASTEER, a large testbed of steered generations, and train a GBDT classifier to predict steering outcomes from early hidden states, enabling efficient steering strength search.

When is Your LLM Steerable?

Hugging Face Daily Papers

This paper introduces a method to predict activation steering effectiveness in language models from early decoding states using a Gradient Boosting Decision Trees (GBDT) classifier, enabling efficient steering strength optimization without full rollouts.