Large language model (LLM)-integrated applications have become increasingly prevalent, yet face critical security vulnerabilities from prompt injection (PI) attacks. Defending against PI attacks faces two major issues: malicious instructions can be injected through diverse vectors, and injected instructions often lack clear semantic boundaries from the surrounding context, making them difficult to identify. To address these issues, we propose InstruCoT, a model enhancement method for PI defense that synthesizes diverse training data and employs instruction-level chain-of-thought fine-tuning, enabling LLMs to effectively identify and reject malicious instructions regardless of their source or position in the context. We evaluate InstruCoT across three critical dimensions: Behavior Deviation, Privacy Leakage, and Harmful Output. Experimental results across four LLMs demonstrate that InstruCoT significantly outperforms baselines in all dimensions while maintaining utility performance without degradation
@article{arxiv.2601.04666,
title = {Know Thy Enemy: Securing LLMs Against Prompt Injection via Diverse Data Synthesis and Instruction-Level Chain-of-Thought Learning},
author = {Zhiyuan Chang and Mingyang Li and Yuekai Huang and Ziyou Jiang and Xiaojun Jia and Qian Xiong and Junjie Wang and Zhaoyang Li and Qing Wang},
journal= {arXiv preprint arXiv:2601.04666},
year = {2026}
}
Comments
19 pages, 6 figures; accepted by ACL 2026 Findings