# VulInstruct: Teaching LLMs Root-Cause Reasoning for Vulnerability Detection via Security Specifications

arXiv: https://arxiv.org/abs/2511.04014 · HTML: https://mlsanity.com/paper/2511.04014/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 25 of its 42 references that are in the map.
- First posted on arXiv 6 Nov 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 25 of 25)

- [A Survey on Code Generation with LLM-based Agents](https://mlsanity.com/paper/2508.00083/) (2025)
- [IRIS: LLM-Assisted Static Analysis for Detecting Security Vulnerabilities](https://mlsanity.com/paper/2405.17238/) (2024)
- [LLMs Cannot Reliably Identify and Reason About Security Vulnerabilities (Yet?): A Comprehensive Evaluation, Framework, and Benchmarks](https://mlsanity.com/paper/2312.12575/) (2023)
- [Vul-RAG: Enhancing LLM-based Vulnerability Detection via Knowledge-level RAG](https://mlsanity.com/paper/2406.11147/) (2024)
- [LLM4Vuln: A Unified Evaluation Framework for Decoupling and Enhancing LLMs' Vulnerability Reasoning](https://mlsanity.com/paper/2401.16185/) (2024)
- [Prompt-Enhanced Software Vulnerability Detection Using ChatGPT](https://mlsanity.com/paper/2308.12697/) (2023)
- [Large Language Model for Vulnerability Detection: Emerging Results and Future Directions](https://mlsanity.com/paper/2401.15468/) (2024)
- [ToolCoder: Teach Code Generation Models to use API search tools](https://mlsanity.com/paper/2305.04032/) (2023)
- [Chain-of-Thought Prompting of Large Language Models for Discovering and Fixing Software Vulnerabilities](https://mlsanity.com/paper/2402.17230/) (2024)
- [Benchmarking LLMs and LLM-based Agents in Practical Vulnerability Detection for Code Repositories](https://mlsanity.com/paper/2503.03586/) (2025)
- [Generalization-Enhanced Code Vulnerability Detection via Multi-Task Instruction Fine-Tuning](https://mlsanity.com/paper/2406.03718/) (2024)
- [Let the Trial Begin: A Mock-Court Approach to Vulnerability Detection using LLM-Based Agents](https://mlsanity.com/paper/2505.10961/) (2025)
- [VulEval: Towards Repository-Level Evaluation of Software Vulnerability Detection](https://mlsanity.com/paper/2404.15596/) (2024)
- [Everything You Wanted to Know About LLM-based Vulnerability Detection But Were Afraid to Ask](https://mlsanity.com/paper/2504.13474/) (2025)
- [ChatCoder: Chat-based Refine Requirement Improves LLMs' Code Generation](https://mlsanity.com/paper/2311.00272/) (2023)
- [aiXcoder-7B-v2: Training LLMs to Fully Utilize the Long Context in Repository-level Code Completion](https://mlsanity.com/paper/2503.15301/) (2025)
- [A Systematic Literature Review on Detecting Software Vulnerabilities with Large Language Models](https://mlsanity.com/paper/2507.22659/) (2025)
- [R2Vul: Learning to Reason about Software Vulnerabilities with Reinforcement Learning and Structured Reasoning Distillation](https://mlsanity.com/paper/2504.04699/) (2025)
- [Context-Enhanced Vulnerability Detection Based on Large Language Model](https://mlsanity.com/paper/2504.16877/) (2025)
- [BugScope: Learn to Find Bugs Like Human](https://mlsanity.com/paper/2507.15671/) (2025)
- [Beyond ChatGPT: Enhancing Software Quality Assurance Tasks with Diverse LLMs and Validation Techniques](https://mlsanity.com/paper/2409.01001/) (2024)
- [Exploring Data-Efficient Adaptation of Large Language Models for Code Generation](https://mlsanity.com/paper/2403.00046/) (2024)
- [If LLMs Would Just Look: Simple Line-by-line Checking Improves Vulnerability Localization](https://mlsanity.com/paper/2410.15288/) (2024)
- [LAMeD: LLM-generated Annotations for Memory Leak Detection](https://mlsanity.com/paper/2505.02376/) (2025)
- [CognixShield: PoV-Guided Vulnerable API Usage Detection in Large Codebases via LLMs](https://mlsanity.com/paper/2506.17798/) (2025)

## Built on by (most cited first, 4 of 4)

- [DREA: Decoupled Reasoning and Exploration Agents for Repository-Level Vulnerability Detection](https://mlsanity.com/paper/2607.13439/) (2026)
- [AEGIS: From Clues to Verdicts -- Graph-Guided Deep Vulnerability Reasoning via Dialectics and Meta-Auditing](https://mlsanity.com/paper/2603.20637/) (2026)
- [LLMs + Security = Trouble](https://mlsanity.com/paper/2602.08422/) (2026)
- [Retrieve, Reproduce, Reveal: Dissecting Retrieval-Augmented Software Vulnerability Detection](https://mlsanity.com/paper/2609.37669/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have achieved remarkable progress in code understanding tasks. However, they demonstrate limited performance in vulnerability detection and struggle to distinguish vulnerable code from patched code. We argue that LLMs lack understanding of security specifications -- the expectations about how code should behave to remain safe. When code behavior differs from these expectations, it becomes a potential vulnerability. However, such knowledge is rarely explicit in training data, leaving models unable to reason about security flaws. We propose VulInstruct, a specification-guided approach that systematically extracts security specifications from historical vulnerabilities to detect new ones. VulInstruct constructs a specification knowledge base from two perspectives: (i) General specifications from high-quality patches across projects, capturing fundamental safe behaviors; and (ii) Domain-specific specifications from repeated violations in particular repositories relevant to the target code. VulInstruct retrieves relevant past cases and specifications, enabling LLMs to reason about expected safe behaviors rather than relying on surface patterns. We evaluate VulInstruct under strict criteria requiring both correct predictions and valid reasoning. On PrimeVul, VulInstruct achieves 45.0% F1-score (32.7% improvement) and 37.7% recall (50.8% improvement) compared to baselines, while uniquely detecting 24.3% of vulnerabilities -- 2.4x more than any baseline. In pair-wise evaluation, VulInstruct achieves 32.3% relative improvement. VulInstruct also discovered a previously unknown high-severity vulnerability (CVE-2025-56538) in production code, demonstrating practical value for real-world vulnerability discovery. All code and supplementary materials are available at https://github.com/zhuhaopku/VulInstruct-temp.