Write a Blog >>

Deliberation is a common and natural behavior in human daily life. For example, when writing papers or articles, we usually first write drafts, and then iteratively polish them until satisfied. In light of such a human cognitive process, we propose DECOM, which is a multi-pass deliberation framework for automatic comment generation. DECOM consists of multiple Deliberation Models and one Evaluation Model. Given a code snippet, we first extract keywords from the code and retrieve a similar code fragment from a pre-defined corpus. Then, we treat the comment of the retrieved code as the initial draft and input it with the code and keywords into DECOM to start the iterative deliberation process. At each deliberation, the deliberation model polishes the draft and generates a new comment. The evaluation model measures the quality of the newly generated comment to determine whether to end the iterative process or not. When the iterative process is terminated, the best-generated comment will be selected as the target comment. Our approach is evaluated on two real-world datasets in Java (87K) and Python (108K), and experiment results show that our approach outperforms the state-of-the-art baselines. A human evaluation study also confirms the comments generated by DECOM tend to be more readable, informative, and useful.

Wed 12 Oct

Displayed time zone: Eastern Time (US & Canada) change

16:00 - 18:00
Technical Session 19 - Formal Methods and Models IResearch Papers / Journal-first Papers / Tool Demonstrations at Ballroom C East
Chair(s): Michalis Famelis Université de Montréal
16:00
20m
Research paper
Automatic Comment Generation via Multi-Pass Deliberation
Research Papers
Fangwen Mu Institute of Software Chinese Academy of Sciences, Xiao Chen Institute of Software Chinese Academy of Sciences, Lin Shi ISCAS, Song Wang York University, Qing Wang Institute of Software at Chinese Academy of Sciences
16:20
10m
Demonstration
Building recommender systems for modelling languages with DroidVirtual
Tool Demonstrations
Lissette Almonte Universidad Autónoma de Madrid, Esther Guerra Universidad Autónoma de Madrid, Iván Cantador Universidad Autónoma de Madrid, Juan de Lara Autonomous University of Madrid
Pre-print Media Attached
16:30
10m
Demonstration
RobSimVer: A Tool for RoboSim Modeling and AnalysisVirtual
Tool Demonstrations
Dehui Du East China Normal University, Ana Cavalcanti University of York, JihuiNie East China Normal University
16:40
20m
Research paper
Provably Tightest Linear Approximation for Robustness Verification of Sigmoid-like Neural NetworksVirtual
Research Papers
Zhaodi Zhang East China Normal University, Yiting Wu East China Normal University, Si Liu ETH Zurich, Jing Liu East China Normal University, Min Zhang East China Normal University
17:00
20m
Research paper
Efficient Synthesis of Method Call Sequences for Test Generation and Bounded VerificationVirtual
Research Papers
Yunfan Zhang Peking University, Ruidong Zhu Peking University, Yingfei Xiong Peking University, Tao Xie Peking University
17:20
20m
Paper
Demystifying Performance Regressions in String SolversVirtual
Journal-first Papers
Yao Zhang , Xiaofei Xie Singapore Management University, Singapore, Yi Li Nanyang Technological University, Yun Lin National University of Singapore, Sen Chen Tianjin University, Yang Liu Nanyang Technological University, Xiaohong Li TianJin University
Link to publication DOI
17:40
20m
Research paper
Detecting Semantic Code Clones by Building AST-based Markov Chains ModelVirtual
Research Papers
Yueming Wu Nanyang Technological University, Siyue Feng Huazhong University of Science and Technology, Deqing Zou Huazhong University of Science and Technology, Hai Jin Huazhong University of Science and Technology