SOM Directions are Better than One: Multi-Directional Refusal Suppression in Language Models

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Piras, Giorgio, Mura, Raffaele, Brau, Fabio, Oneto, Luca, Roli, Fabio, Biggio, Battista
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918406886785024
author Piras, Giorgio
Mura, Raffaele
Brau, Fabio
Oneto, Luca
Roli, Fabio
Biggio, Battista
author_facet Piras, Giorgio
Mura, Raffaele
Brau, Fabio
Oneto, Luca
Roli, Fabio
Biggio, Battista
contents Refusal refers to the functional behavior enabling safety-aligned language models to reject harmful or unethical prompts. Following the growing scientific interest in mechanistic interpretability, recent work encoded refusal behavior as a single direction in the model's latent space; e.g., computed as the difference between the centroids of harmful and harmless prompt representations. However, emerging evidence suggests that concepts in LLMs often appear to be encoded as a low-dimensional manifold embedded in the high-dimensional latent space. Motivated by these findings, we propose a novel method leveraging Self-Organizing Maps (SOMs) to extract multiple refusal directions. To this end, we first prove that SOMs generalize the prior work's difference-in-means technique. We then train SOMs on harmful prompt representations to identify multiple neurons. By subtracting the centroid of harmless representations from each neuron, we derive a set of multiple directions expressing the refusal concept. We validate our method on an extensive experimental setup, demonstrating that ablating multiple directions from models' internals outperforms not only the single-direction baseline but also specialized jailbreak algorithms, leading to an effective suppression of refusal. Finally, we conclude by analyzing the mechanistic implications of our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2511_08379
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SOM Directions are Better than One: Multi-Directional Refusal Suppression in Language Models
Piras, Giorgio
Mura, Raffaele
Brau, Fabio
Oneto, Luca
Roli, Fabio
Biggio, Battista
Artificial Intelligence
Machine Learning
Refusal refers to the functional behavior enabling safety-aligned language models to reject harmful or unethical prompts. Following the growing scientific interest in mechanistic interpretability, recent work encoded refusal behavior as a single direction in the model's latent space; e.g., computed as the difference between the centroids of harmful and harmless prompt representations. However, emerging evidence suggests that concepts in LLMs often appear to be encoded as a low-dimensional manifold embedded in the high-dimensional latent space. Motivated by these findings, we propose a novel method leveraging Self-Organizing Maps (SOMs) to extract multiple refusal directions. To this end, we first prove that SOMs generalize the prior work's difference-in-means technique. We then train SOMs on harmful prompt representations to identify multiple neurons. By subtracting the centroid of harmless representations from each neuron, we derive a set of multiple directions expressing the refusal concept. We validate our method on an extensive experimental setup, demonstrating that ablating multiple directions from models' internals outperforms not only the single-direction baseline but also specialized jailbreak algorithms, leading to an effective suppression of refusal. Finally, we conclude by analyzing the mechanistic implications of our approach.
title SOM Directions are Better than One: Multi-Directional Refusal Suppression in Language Models
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2511.08379