7 papers
ECLAIR: A Causally-Grounded AI Framework for Scientific Discovery in Empirical Software Engineering
Alejandro Velasco, Daniel Rodriguez-Cardenas, Dipin Khati +2
The scientific method has long guided empirical research in Software Engineering (SE), but the complexity of modern software systems often hinders its systematic application. This…
Enabling Global, Human-Centered Explanations for LLMs:From Tokens to Interpretable Code and Test Generation
Dipin Khati, Daniel Rodriguez-Cardenas, David N. Palacio +3
As Large Language Models for Code (LM4Code) become integral to software engineering, establishing trust in their output becomes critical. However, standard accuracy metrics obscure…
Towards Comprehensive Benchmarking Infrastructure for LLMs In Software Engineering
Daniel Rodriguez-Cardenas, Xiaochang Li, Marcos Macedo +5
Large language models for code are advancing fast, yet our ability to evaluate them lags behind. Current benchmarks focus on narrow tasks and single metrics, which hide critical ga…
Detecting and Correcting Hallucinations in LLM-Generated Code via Deterministic AST Analysis
Dipin Khati, Daniel Rodriguez-Cardenas, Paul Pantzer +1
Large Language Models (LLMs) for code generation boost productivity but frequently introduce Knowledge Conflicting Hallucinations (KCHs), subtle, semantic errors, such as non-exist…
Tricky: Towards a Benchmark for Evaluating Human and LLM Error Interactions
Cole Granger, Dipin Khati, Daniel Rodriguez-Cardenas +1
Large language models (LLMs) are increasingly integrated into software development workflows, yet they often introduce subtle logic or data-misuse errors that differ from human bug…
A Causal Perspective on Measuring, Explaining and Mitigating Smells in LLM-Generated Code
Alejandro Velasco, Daniel Rodriguez-Cardenas, Dipin Khati +3
Recent advances in large language models (LLMs) have accelerated their adoption in software engineering contexts. However, concerns persist about the structural quality of the code…