Pruning as a Defense: Reducing Memorization in Large Language Models

2025-02-18

AI Summary

Large language models (LLMs) inherently memorize training data, posing significant privacy risks—including data leakage and membership inference attacks. This work systematically demonstrates, for the first time, that structured pruning and sparsification—guided by weight importance estimation—effectively suppress LLMs’ memorization behavior, establishing “pruning-as-defense” as a novel privacy-preserving paradigm. Evaluated across multiple mainstream LLMs, our approach reduces memory leakage rates by 40–65% while retaining over 90% of original task performance. Crucially, experiments conducted under standardized membership inference attack benchmarks confirm that pruning is not merely a model compression technique but serves as a lightweight, general-purpose, and retraining-free foundational defense mechanism for privacy enhancement.

Technology Category

• Machine Learning: Privacy
• Natural Language Processing: (Large) Language Models
• Computer Vision: Large Vision Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Security and Privacy: Privacy-enhancing technologies
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Large language models have been shown to memorize significant portions of their training data, which they can reproduce when appropriately prompted. This work investigates the impact of simple pruning techniques on this behavior. Our findings reveal that pruning effectively reduces the extent of memorization in LLMs, demonstrating its potential as a foundational approach for mitigating membership inference attacks.

Problem

• Reducing memorization in LLMs
• Impact of pruning techniques
• Mitigating membership inference attacks

Innovation

• Pruning reduces memorization
• Technique mitigates inference attacks
• Foundational approach for LLMs