Share your thoughts, 1 month free Claude Pro on usSee more

Code Summarization Factual Consistency on Python

0.497Pearson Correlation (rp)

ReFEree

Updated 4d ago

Evaluation Results

Method	Links
ReFEree 2026.04		0.497	0.489	0.39	0.459
ReFEree 2026.04		0.432	0.432	0.349	0.404
G-Eval 2026.04		0.427	0.413	0.36	0.4
LLM-judge 2026.04		0.419	0.41	0.36	0.396
CODERPE 2026.04		0.418	0.405	0.353	0.392
Factscore 2026.04		0.41	0.426	0.338	0.391
SBCS 2026.04		0.068	0.036	0.028	0.044
ROUGE-1 2026.04		0.046	0.044	0.034	0.041
ROUGE-L 2026.04		0.045	0.037	0.029	0.037
METEOR 2026.04		0.033	0.024	0.019	0.025
ROUGE-2 2026.04		0.03	0.031	0.024	0.028
BLEU 2026.04		0.016	0.018	0.015	0.017
BERTScore 2026.04		0.015	0	0	0.005
SBED 2026.04		-0.058	-0.036	-0.028	-0.041
SIDE 2026.04		-0.064	-0.056	-0.043	-0.054