Fine-tuned GPT-4o classifies GitHub issue types with about 86% F1 on NLBSE 2024 data, while a much larger NLBSE 2023 dataset does not improve results.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Applying Large Language Models to Issue Classification: Revisiting with Extended Data and New Models
Fine-tuned GPT-4o classifies GitHub issue types with about 86% F1 on NLBSE 2024 data, while a much larger NLBSE 2023 dataset does not improve results.