Natural Code Transformation Confuses Agents in Vulnerability Analysis
cs.CR, cs.AI, cs.LG
Submitted: 2026-10-09
Updated: 2026-10-14
Code: https://github.com/cesanta/mon
Sources
- DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
- Program Analysis Guided LLM Agent for Proof-of-Concept Generation
- GLM-5: from Vibe Coding to Agentic Engineering
- RepoAudit: An Autonomous LLM-Agent for Repository-Level Code Auditing
- Capture the Flags: Family-Based Evaluation of Agentic LLMs via Semantics-Preserving Transformations
- Auto-SPT: Automating Semantic Preserving Transformations for Code
- On the Adversarial Robustness of Instruction-Tuned Large Language Models for Code
- SoK: Agentic Skills -- Beyond Tool Use in LLM Agents
- ExploitBench: A Capability Ladder Benchmark for LLM Cybersecurity Agents
- SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks
- A Systematic Study of Code Obfuscation Against LLM-based Vulnerability Detection
- Exploiting Code Symmetries for Learning Program Semantics
- PatchBench: Evaluating AI Agents for Vulnerability Patching
- Intriguing properties of neural networks
- ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks?
- Symbolon: Symbolic Execution by Learning Code Transformation
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs