4 papers
Authority Is Not a String: A Capability-Scoped Harness for Prompt-Injection-Resistant Coding Agents
Dimitrios Stamatios Bouras, Yihan Dai, Sergey Mechtaev
Coding agents use system-level tools to read files, execute commands, and modify source code. Within the agent's sandbox, these tools often carry ambient authority: naming a resour…
Compressing Code Context for LLM-based Issue Resolution
Haoxiang Jia, Earl T. Barr, Sergey Mechtaev
Large Language Models (LLMs) are now capable of resolving real-world GitHub issues. However, current approaches overapproximate the code context and suffer from two compounding pro…
Defusing Logic Bombs in Symbolic Execution with LLM-Generated Ghost Code
Dimitrios Stamatios Bouras, Sergey Mechtaev
Symbolic execution is a powerful program analysis technique, but its effectiveness is fundamentally limited by solver-hostile program fragments, complex numerical reasoning, and un…
Reducing Hallucinations in LLM-Generated Code via Semantic Triangulation
Yihan Dai, Sijie Liang, Haotian Xu +2
Large language models (LLMs) can generate executable code from natural language descriptions, but the resulting programs frequently contain bugs due to hallucinations. In the absen…