DPN-LE: Dual Personality Neuron Localization and Editing for Large Language Models

arXiv cs.CL / 5/1/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces a new approach to personality editing in LLMs, arguing that prior neuron-editing methods often change overall performance because they modify many neurons not truly tied to personality.
It empirically finds that LLM neurons are multifunctional—linking personality traits with general knowledge—and that neurons representing opposing traits show mutually exclusive representation patterns.
Based on these findings, the authors propose DPN-LE, which identifies personality-specific neurons by contrasting MLP activations from high-trait vs. low-trait samples.
DPN-LE builds layer-wise steering vectors and uses dual-criterion filtering (Cohen’s d effect size and activation magnitude) to select small, trait-specific neuron subsets and then applies sparse interventions at inference time.
With only 1,000 contrastive sample pairs per trait, DPN-LE edits about 0.5% of neurons while maintaining competitive personality control and substantially better capability preservation on reasoning tasks, validated on LLaMA-3-8B-Instruct and Qwen2.5-7B-Instruct.

Abstract

With the widespread adoption of large language models (LLMs), understanding their personality representation mechanisms has become critical. As a novel paradigm in Personality Editing, most existing methods employ neuron-editing to locate and modify LLM neurons, requiring changes to numerous neurons and leading to significant performance degradation. This raises a fundamental question: Are all modified neurons directly related to personality representation? In this work, we investigate and quantify this specificity through assessments of general capability impact and representation-level patterns. We find that: 1) Current methods can change personalities but reduce overall performance. 2) Neurons are multifunctional, connecting personality traits and general knowledge. 3) Opposing personality traits demonstrate distinctly mutually exclusive representation patterns. Motivated by these findings, we propose DPN-LE (Dual Personality Neuron Localization and Editing), which identifies personality-specific neurons by contrasting MLP activations between high-trait and low-trait samples. DPN-LE constructs layer-wise steering vectors and applies dual-criterion filtering based on Cohen's

d

effect size and activation magnitude to isolate mutually exclusive neuron subsets. Sparse linear intervention on these neurons enables precise personality control at inference time. Using only 1,000 contrastive sample pairs per trait, DPN-LE intervenes on

\sim

0.5\% of neurons while achieving competitive personality control and substantially better capability preservation across reasoning tasks. Experiments on LLaMA-3-8B-Instruct and Qwen2.5-7B-Instruct demonstrate the effectiveness and generalizability of our approach.