3 papers
cs.SE2026
Diff-Based Code Corruption using LLMs for Large-Scale Bugfix Benchmarking
Balázs Szalontai, Ãbel Szauter, Balázs Márton +3
There are various benchmarks to evaluate bugfixing capabilities of Large Language Models. However, most widespread benchmarks do not fully reflect real-world bugfixing practices. T…
cs.LG2026
Neural architectures for resolving references in program code
GergŠSzalay, Gergely Zsolt Kovács, Sándor Teleki +2
Resolving and rewriting references is fundamental in programming languages. Motivated by a real-world decompilation task, we abstract reference rewriting into the problems of direc…
cs.AI2024
Large Language Models for Code Summarization
Balázs Szalontai, GergŠSzalay, Tamás Márton +3
Recently, there has been increasing activity in using deep learning for software engineering, including tasks like code generation and summarization. In particular, the most recent…