Factual Consistency Evaluation on 2,055 code summary evaluations benchmark leaderboard · SOTA2 Research

Factual Consistency Evaluation on 2,055 code summary evaluations

0Time (s)ROUGE-1

Evaluation Results

Method	Links
ROUGE-12026.04		0	0
ROUGE-22026.04		0	0
ROUGE-L2026.04		0	0
METEOR2026.04		0.02	0
SBCS2026.04		0.02	0
BERTScore2026.04		0.03	0
SBED2026.04		0.03	0
SIDE2026.04		0.08	0
LLM-judge2026.04		0.34	0.0002
CODERPE2026.04		0.53	0.0002
G-Eval2026.04		1.11	0.0002
BLEU2026.04		1.12	0
Factscore2026.04		9.18	0.0058
ReFEree2026.04		10.24	0.0042