Asking an LLM to rate a code summary directly correlates with human judgments as well as or better than n-gram, learned, and embedding metrics, even without a reference summary.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Simple and Effective Baselines for Code Summarisation Evaluation
Asking an LLM to rate a code summary directly correlates with human judgments as well as or better than n-gram, learned, and embedding metrics, even without a reference summary.