A feature-model-driven LLM pipeline that generates synthetic requirements data improves defect classification when combined with real data, but the headline gains rest on a 40-sample test set with high variance.
Few-shot fine-tuning vs. in-context learning: A fair comparison and evaluation
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
-
Synthline: A Product Line Approach for Synthetic Requirements Engineering Data Generation using Large Language Models
A feature-model-driven LLM pipeline that generates synthetic requirements data improves defect classification when combined with real data, but the headline gains rest on a 40-sample test set with high variance.