Precision Knowledge Editing: Enhancing Safety in Large Language Models

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Xuying, Li, Zhuo, Kosuga, Yuji, Yoshida, Yasuhiro, Bian, Victor
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929537426653184
author Li, Xuying
Li, Zhuo
Kosuga, Yuji
Yoshida, Yasuhiro
Bian, Victor
author_facet Li, Xuying
Li, Zhuo
Kosuga, Yuji
Yoshida, Yasuhiro
Bian, Victor
contents Large language models (LLMs) have demonstrated remarkable capabilities, but they also pose risks related to the generation of toxic or harmful content. This work introduces Precision Knowledge Editing (PKE), an advanced technique that builds upon existing knowledge editing methods to more effectively identify and modify toxic parameter regions within LLMs. By leveraging neuron weight tracking and activation pathway tracing, PKE achieves finer granularity in toxic content management compared to previous methods like Detoxifying Instance Neuron Modification (DINM). Our experiments demonstrate that PKE significantly reduces the attack success rate (ASR) across various models, including Llama2-7b and Llama-3-8b-instruct, while maintaining overall model performance. Additionally, we also compared the performance of some closed-source models (gpt-4-0613 and Claude 3 Sonnet) in our experiments, and found that models adjusted using our method far outperformed the closed-source models in terms of safety. This research contributes to the ongoing efforts to make LLMs safer and more reliable for real-world applications.
format Preprint
id arxiv_https___arxiv_org_abs_2410_03772
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Precision Knowledge Editing: Enhancing Safety in Large Language Models
Li, Xuying
Li, Zhuo
Kosuga, Yuji
Yoshida, Yasuhiro
Bian, Victor
Computation and Language
Artificial Intelligence
Large language models (LLMs) have demonstrated remarkable capabilities, but they also pose risks related to the generation of toxic or harmful content. This work introduces Precision Knowledge Editing (PKE), an advanced technique that builds upon existing knowledge editing methods to more effectively identify and modify toxic parameter regions within LLMs. By leveraging neuron weight tracking and activation pathway tracing, PKE achieves finer granularity in toxic content management compared to previous methods like Detoxifying Instance Neuron Modification (DINM). Our experiments demonstrate that PKE significantly reduces the attack success rate (ASR) across various models, including Llama2-7b and Llama-3-8b-instruct, while maintaining overall model performance. Additionally, we also compared the performance of some closed-source models (gpt-4-0613 and Claude 3 Sonnet) in our experiments, and found that models adjusted using our method far outperformed the closed-source models in terms of safety. This research contributes to the ongoing efforts to make LLMs safer and more reliable for real-world applications.
title Precision Knowledge Editing: Enhancing Safety in Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2410.03772