SOTAVerified|Agents Browse Leaderboard About

Source Code Summarization

Code Summarization is a task that tries to comprehend code and automatically generate descriptions directly from the source code.

Source: Improving Automatic Source Code Summarization via Deep Reinforcement Learning

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 31–40 of 58 papers

Title	Date	Tasks	Status	Hype	Score
Tram: A Token-level Retrieval-augmented Mechanism for Source Code Summarization	May 18, 2023	Code SummarizationDecoder	CodeCode Available	0	5
Variational Prefix Tuning for Diverse and Accurate Code Summarization Using Pre-trained Language Models	May 14, 2025	Code SummarizationReranking	CodeCode Available	0	5
Automatic Source Code Summarization with Extended Tree-LSTM	Jun 19, 2019	Code SummarizationMachine Translation	CodeCode Available	0	5
code2seq: Generating Sequences from Structured Representations of Code	Aug 4, 2018	Code SummarizationNMT	CodeCode Available	0	5
GraphCodeBERT: Pre-training Code Representations with Data Flow	Sep 17, 2020	Clone DetectionCode Completion	CodeCode Available	0	5
HAConvGNN: Hierarchical Attention Based Convolutional Graph Neural Network for Code Documentation Generation in Jupyter Notebooks	Mar 31, 2021	Code Documentation GenerationCode Summarization	CodeCode Available	0	5
Improving Automatic Source Code Summarization via Deep Reinforcement Learning	Nov 17, 2018	Code SummarizationDecoder	CodeCode Available	0	5
Code Generation as a Dual Task of Code Summarization	Oct 14, 2019	Code GenerationCode Summarization	CodeCode Available	0	5
CoDesc: A Large Code-Description Parallel Dataset	May 29, 2021	Code SearchCode Summarization	CodeCode Available	0	5
Meta Learning for Code Summarization	Nov 16, 2021	Code SummarizationMeta-Learning	—Unverified	0	0

Show:10 25 50

← PrevPage 4 of 6Next →

All datasets DeepCom-Java ParallelCorpus-Python CoDesc CodeSearchNet CodeSearchNet - Python Java scripts Summarizing Source Code using a Neural Attention Model - C#Summarizing Source Code using a Neural Attention Model - Python Summarizing Source Code using a Neural Attention Model - SQL

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	AdaMo-noise	BLEU-4	45.35	—	Unverified
2	AdaMo-basic	BLEU-4	45.3	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	AdaMo-noise	BLEU-4	34.05	—	Unverified
2	AdaMo-basic	BLEU-4	33.85	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Transformer	BLEU-4	45.89	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ContraCode	F1	17.24	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	AdaMo-basic	BLEU-4	16.46	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	AdaMo-basic	BLEU-4	37.64	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	CodeTrans-MT-Large	Smoothed BLEU-4	23.57	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	CodeTrans-MT-Base	Smoothed BLEU-4	13.37	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	CodeTrans-MT-TF-Large	Smoothed BLEU-4	19.98	—	Unverified