Foutse Khomh
5 papers · Latest:
A Research Agenda on Agents and Software Engineering: Outcomes from the Rio A2SE Seminar
This paper outlines a community-driven research agenda for agents and software engineering, covering six key thematic areas identified by experts.
On Fixing Insecure AI-Generated Code through Model Fine-Tuning and Prompting Strategies
This paper systematically investigates strategies for fixing insecure AI-generated code, revealing no universal solution and high dependency on strategy/model.
Mitigating False Positives in Static Memory Safety Analysis of Rust Programs via Reinforcement Learning
This paper uses reinforcement learning with dynamic fuzzing to significantly reduce false positives in static memory safety analysis of Rust programs.
Self-Adaptive Multi-Agent LLM-Based Security Pattern Selection for IoT Systems
ASPO uses multi-agent LLMs and deterministic optimization for self-adaptive, resource-efficient security pattern selection in IoT edge systems.
Structural Anchors and Reasoning Fragility:Understanding CoT Robustness in LLM4Code
CoT robustness in LLM4Code is not uniform; its benefits depend on model, task, prompt, and how perturbations affect structural anchors.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.