• Home
  • Search
  • CoSS: leveraging statement semantics for code summarization
  • Cite Icon6
  • https://doi.org/10.1109/tse.2023.3256362Copy DOI Icon

CoSS: leveraging statement semantics for code summarization

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Automated code summarization tools allow generating descriptions for code snippets in natural language, which benefits software development and maintenance. Recent studies demonstrate that the quality of generated summaries can be improved by using additional code representations beyond token sequences. The majority of contemporary approaches mainly focus on extracting code syntactic and structural information from abstract syntax trees (ASTs). However, from the view of macro-structures, it is challenging to identify and capture semantically meaningful features due to fine-grained syntactic nodes involved in ASTs. To fill this gap, we investigate how to learn more code semantics and control flow features from the perspective of code statements. Accordingly, we propose a novel model entitled CoSS for code summarization. CoSS adopts a Transformer-based encoder and a graph attention network-based encoder to capture token-level and statement-level semantics from code token sequence and control flow graph, respectively. Then, after receiving two-level embeddings from encoders, a joint decoder with a multi-head attention mechanism predicts output sequences verbatim. Performance evaluations on Java, Python, and Solidity datasets validate that CoSS outperforms nine state-of-the-art (SOTA) neural code summarization models in effectiveness and is competitive in execution efficiency. Further, the ablation study reveals the contribution of each model component.

Similar Papers
  • Conference Article
  • Citations78

Leveraging Code Generation to Improve Code Retrieval and Summarization via Dual Learning

  • Apr 20, 2020
  • Wei Ye +5
  • Research Article
  • Citations11

Do Code Summarization Models Process Too Much Information? Function Signature May Be All That Is Needed

  • Jun 27, 2024
  • ACM Transactions on Software Engineering and Methodology
  • Xi Ding +5
  • Conference Article
  • Citations2

Abstract Syntax Tree (AST) and Control Flow Graph (CFG) Construction of Notasi Algoritmik

  • Nov 03, 2021
  • Irfan Sofyana Putra +2
  • Video Transcripts

CoDesc: A Large Code–Description Parallel Dataset

  • Aug 03, 2021
  • Underline Science Inc.
  • Masum Hasan +1
  • Research Article
  • Citations4

On the impact of multiple source code representations on software engineering tasks — An empirical study

  • Dec 27, 2023
  • Journal of Systems and Software
  • Karthik Chandra Swarna +3
  • Conference Article
  • Citations2

BCFA

  • Jun 27, 2020
  • Ramanathan Ramu +3
  • Conference Article
  • Citations1

Handling context-sensitive syntactic issues in the design of a front-end for a MATLAB compiler

  • Jul 24, 2000
  • Pramod G Joisha +4
  • Conference Article
  • Citations22

Adaptive Call-Site Sensitive Control Flow Integrity

  • Jun 01, 2019
  • Mustakimur Khandaker +5
  • Dissertation

Trust, transforms, and control flow: A graph-theoretic method to verifying source and binary control flow equivalence

  • Jan 01, 2021
  • Ryan Christopher Goluch
  • Conference Article
  • Citations2

MazeRunner: Evaluating the Attack Surface of Control-Flow Integrity Policies

  • Oct 01, 2021
  • Dongrui Zeng +2
  • Video Transcripts

Unified Pre-training for Program Understanding and Generation

  • May 25, 2021
  • Underline Science Inc.
  • Saikat Chakraborty
  • Conference Article
  • Citations10

GrammarT5: Grammar-Integrated Pretrained Encoder-Decoder Neural Model for Code

  • Apr 12, 2024
  • Qihao Zhu +5
  • Research Article
  • Citations15

The flowing nature matters: feature learning from the control flow graph of source code for bug localization

  • Feb 17, 2022
  • Machine Learning
  • Yi-Fan Ma +1
  • Conference Article

Improve Neural Machine Translation by Syntax Tree

  • Sep 21, 2018
  • Siyu Chen +1
  • Research Article

Cloud-assisted LLM-enhanced datasets for AST hierarchy-aware code summarization model

  • Feb 14, 2026
  • Journal of Cloud Computing
  • Junsan Zhang +5
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.