# From Trace to Line: LLM Agent for Real-World OSS Vulnerability Localization

arXiv: https://arxiv.org/abs/2510.02389 · HTML: https://mlsanity.com/paper/2510.02389/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 34 of its 60 references that are in the map.
- First posted on arXiv 30 Sep 2025.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 30 of 34)

- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [A Survey on the Memory Mechanism of Large Language Model based Agents](https://mlsanity.com/paper/2404.13501/) (2024)
- [From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judge](https://mlsanity.com/paper/2411.16594/) (2024)
- [Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks](https://mlsanity.com/paper/1909.03496/) (2019)
- [InterCode: Standardizing and Benchmarking Interactive Coding with Execution Feedback](https://mlsanity.com/paper/2306.14898/) (2023)
- [Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models](https://mlsanity.com/paper/2408.08926/) (2024)
- [NYU CTF Bench: A Scalable Open-Source Benchmark Dataset for Evaluating LLMs in Offensive Security](https://mlsanity.com/paper/2406.05590/) (2024)
- [When LLMs Meet Cybersecurity: A Systematic Literature Review](https://mlsanity.com/paper/2405.03644/) (2024)
- [Large Language Models for Test-Free Fault Localization](https://mlsanity.com/paper/2310.01726/) (2023)
- [ProtAgents: Protein discovery via large language model multi-agent collaborations combining physics and machine learning](https://mlsanity.com/paper/2402.04268/) (2024)
- [EnIGMA: Interactive Tools Substantially Assist LM Agents in Finding Security Vulnerabilities](https://mlsanity.com/paper/2409.16165/) (2024)
- [Data Quality for Software Vulnerability Datasets](https://mlsanity.com/paper/2301.05456/) (2023)
- [A Quantitative and Qualitative Evaluation of LLM-Based Explainable Fault Localization](https://mlsanity.com/paper/2308.05487/) (2023)
- [AgentFL: Scaling LLM-based Fault Localization to Project-Level Context](https://mlsanity.com/paper/2403.16362/) (2024)
- [PentestAgent: Incorporating LLM Agents to Automated Penetration Testing](https://mlsanity.com/paper/2411.05185/) (2024)
- [An Empirical Evaluation of LLMs for Solving Offensive Security Challenges](https://mlsanity.com/paper/2402.11814/) (2024)
- [AutoSafeCoder: A Multi-Agent Framework for Securing LLM Code Generation through Static Analysis and Fuzz Testing](https://mlsanity.com/paper/2409.10737/) (2024)
- [Survey of different Large Language Model Architectures: Trends, Benchmarks, and Challenges](https://mlsanity.com/paper/2412.03220/) (2024)
- [LLMs and the Future of Chip Design: Unveiling Security Risks and Building Trust](https://mlsanity.com/paper/2405.07061/) (2024)
- [D-CIPHER: Dynamic Collaborative Intelligent Multi-Agent System with Planner and Heterogeneous Executors for Offensive Security](https://mlsanity.com/paper/2502.10931/) (2025)
- [ChemGraph: An Agentic Framework for Computational Chemistry Workflows](https://mlsanity.com/paper/2506.06363/) (2025)
- [CRAKEN: Cybersecurity LLM Agent with Knowledge-Based Execution](https://mlsanity.com/paper/2505.17107/) (2025)
- [ProLLM: Protein Chain-of-Thoughts Enhanced LLM for Protein-Protein Interaction Prediction](https://mlsanity.com/paper/2405.06649/) (2024)
- [LLMs for Cyber Security: New Opportunities](https://mlsanity.com/paper/2404.11338/) (2024)
- [From Promise to Peril: Rethinking Cybersecurity Red and Blue Teaming in the Age of LLMs](https://mlsanity.com/paper/2506.13434/) (2025)
- [A Multi-Agent Approach to Fault Localization via Graph-Based Retrieval and Reflexion](https://mlsanity.com/paper/2409.13642/) (2024)
- [A PennyLane-Centric Dataset to Enhance LLM-based Quantum Code Generation using RAG](https://mlsanity.com/paper/2503.02497/) (2025)
- [Large Language Models for In-File Vulnerability Localization Can Be "Lost in the End"](https://mlsanity.com/paper/2502.06898/) (2025)
- [Benchmarking LLMs in an Embodied Environment for Blue Team Threat Hunting](https://mlsanity.com/paper/2505.11901/) (2025)
- [Improving LLM-Based Fault Localization with External Memory and Project Context](https://mlsanity.com/paper/2506.03585/) (2025)

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

- [CTFExplorer: Evaluating LLM Offensive Agents Through Multi-Target Web CTF Benchmarking](https://mlsanity.com/paper/2602.08023/) (2026)
- [Antares: Foundation Models for Agentic Vulnerability Localization](https://mlsanity.com/paper/2608.02407/) (2026)
- [AI In Cybersecurity Education -- Scalable Agentic CTF Design Principles and Educational Outcomes](https://mlsanity.com/paper/2603.21551/) (2026)
- [RAVEN: Retrieval-Augmented Vulnerability Exploration Network for Memory Corruption Analysis in User Code and Binary Programs](https://mlsanity.com/paper/2604.17948/) (2026)

## Abstract (arXiv)

Large language models show promise for vulnerability discovery, yet prevailing methods inspect code in isolation, struggle with long contexts, and focus on coarse function- or file-level detections that offer limited guidance to engineers who need precise line-level localization for targeted patches. We introduce T2L, an executable framework for project-level, line-level vulnerability localization that progressively narrows scope from repository modules to exact vulnerable lines via AST-based chunking and evidence-guided refinement. We provide a baseline agent with an Agentic Trace Analyzer (ATA) that fuses runtime evidence such as crash points and stack traces to translate failure symptoms into actionable diagnoses. To enable rigorous evaluation, we introduce T2L-ARVO, an expert-verified 50-case benchmark spanning five crash families in real-world projects. On T2L-ARVO, our baseline achieves up to 58.0% detection and 54.8% line-level localization rate. Together, T2L framework advance LLM-based vulnerability detection toward deployable, precision diagnostics in open-source software workflows.