Cross-Task Defense: Instruction-Tuning LLMs for Content Safety
Yu Fu, Wen Xiao, Jia Chen, Jiachen Li, Evangelos Papalexakis, Aichi Chien, Yue Dong
Abstract
Recent studies reveal that Large Language Models (LLMs) face challenges in balancing safety with utility, particularly when processing long texts for NLP tasks like summarization and translation. Despite defenses against malicious short questions, the ability of LLMs to safely handle dangerous long content, such as manuals teaching illicit activities, remains unclear. Our work aims to develop robust defenses for LLMs in processing malicious documents alongside benign NLP task queries. We introduce a defense dataset comprised of safety-related examples and propose single-task and mixed-task losses for instruction tuning. Our empirical results demonstrate that LLMs can significantly enhance their capacity to safely manage dangerous content with appropriate instruction tuning. Additionally, strengthening the defenses of tasks most susceptible to misuse is effective in protecting LLMs against processing harmful information. We also observe that trade-offs between utility and safety exist in defense strategies, where Llama2, utilizing our proposed approach, displays a significantly better balance compared to Llama1.- Anthology ID:
- 2024.trustnlp-1.9
- Volume:
- Proceedings of the 4th Workshop on Trustworthy Natural Language Processing (TrustNLP 2024)
- Month:
- June
- Year:
- 2024
- Address:
- Mexico City, Mexico
- Editors:
- Kai-Wei Chang, Anaelia Ovalle, Jieyu Zhao, Yang Trista Cao, Ninareh Mehrabi, Aram Galstyan, Jwala Dhamala, Anoop Kumar, Rahul Gupta
- Venues:
- TrustNLP | WS
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 85–93
- Language:
- URL:
- https://aclanthology.org/2024.trustnlp-1.9
- DOI:
- Cite (ACL):
- Yu Fu, Wen Xiao, Jia Chen, Jiachen Li, Evangelos Papalexakis, Aichi Chien, and Yue Dong. 2024. Cross-Task Defense: Instruction-Tuning LLMs for Content Safety. In Proceedings of the 4th Workshop on Trustworthy Natural Language Processing (TrustNLP 2024), pages 85–93, Mexico City, Mexico. Association for Computational Linguistics.
- Cite (Informal):
- Cross-Task Defense: Instruction-Tuning LLMs for Content Safety (Fu et al., TrustNLP-WS 2024)
- PDF:
- https://preview.aclanthology.org/jeptaln-2024-ingestion/2024.trustnlp-1.9.pdf