Does In-IDE Calibration of Large Language Models Work at Scale?
Empirical investigation of calibration techniques for in-IDE LLM suggestions at scale.
Code4MeV2: A Research-oriented Code-completion Platform
Platform for code-completion research combining IDE plugin, backend, and ML infrastructure.
Are Agents Just Automata? On the Formal Equivalence Between Agentic AI and the Chomsky Hierarchy
Theoretical work exploring formal models of agentic AI via classical automata and language hierarchies.
AgentGuard: Runtime Verification of AI Agents
Runtime verification framework to ensure AI agents operate within defined constraints and enable auditing.
Benchmarking AI Models in Software Engineering: A Review, Search Tool, and Enhancement Protocol
Review and tooling for elevating benchmark quality in AI4SE; introduces BenchScout and an enhancement protocol.
Leveraging Large Language Models for Enhancing the Understandability of Generated Unit Tests
Preprint on improving developer understanding of LLM-generated unit tests.
Amirhossein Deljouyi