SEQUENCER: Sequence-to-Sequence Learning for End-to-End Program RepairJ1
This paper presents a novel end-to-end approach to program repair based on sequence-to-sequence learning. We devise, implement, and evaluate a technique, called SEQUENCER, for fixing bugs based on sequence-to-sequence learning on source code. This approach uses the copy mechanism to overcome the unlimited vocabulary problem that occurs with big code. Our system is data-driven; we train it on 35,578 samples, carefully curated from commits to open-source repositories. We evaluate SEQUENCER on 4,711 independent real bug fixes, as well on the Defects4J benchmark used in program repair research. SEQUENCER is able to perfectly predict the fixed line for 950/4,711 testing samples, and find correct patches for 14 bugs in Defects4J benchmark. SEQUENCER captures a wide range of repair operators without any domain-specific top-down design.
Wed 8 JulDisplayed time zone: (UTC) Coordinated Universal Time change
00:00 - 01:00 | P9-Bugs and RepairJournal First / Technical Papers / Software Engineering in Practice / New Ideas and Emerging Results at Silla Chair(s): Yingfei Xiong Peking University, China | ||
00:00 12mTalk | PRECFIX: Large-Scale Patch Recommendation by Mining Defect-Patch PairsSEIP Software Engineering in Practice Xindong Zhang Alibaba Group, Chenguang Zhu University of Texas, Austin, Yi Li Nanyang Technological University, Jianmei Guo Alibaba Group, Lihua Liu Alibaba Group, Haobo Gu Alibaba Group Pre-print | ||
00:12 12mTalk | On the Efficiency of Test Suite based Program Repair: A Systematic Assessment of 16 Automated Repair Systems for Java Programs Technical Papers Kui Liu Huawei Software Engineering Application Technology Lab, Shangwen Wang National University of Defense Technology, Anil Koyuncu University of Luxembourg, Luxembourg, Kisub Kim University of Luxembourg, SnT, Tegawendé F. Bissyandé SnT, University of Luxembourg, Dongsun Kim Furiosa.ai, Peng Wu National University of Defense Technology, Jacques Klein University of Luxembourg, SnT, Xiaoguang Mao National University of Defense Technology, Yves Le Traon University of Luxembourg Pre-print | ||
00:24 8mTalk | SEQUENCER: Sequence-to-Sequence Learning for End-to-End Program RepairJ1 Journal First Zimin Chen KTH Royal Institute of Technology, Steve Kommrusch Colorado State University, Michele Tufano College of William and Mary, Louis-Noël Pouchet Colorado State University, USA, Denys Poshyvanyk William and Mary, Martin Monperrus KTH Royal Institute of Technology | ||
00:32 8mTalk | A Study of Bug Resolution Characteristics in Popular Programming LanguagesJ1 Journal First Jie M. Zhang University College London, UK, Feng Li , Dan Hao Peking University, Meng Wang University of Bristol, UK, Hao Tang Peking University, Lu Zhang Peking University, China, Mark Harman Facebook and University College London | ||
00:40 12mTalk | Automated Bug Reproduction from User Reviews for Android ApplicationsSEIP Software Engineering in Practice Shuyue Li Xi'an Jiaotong University, Jiaqi Guo Xi'an Jiaotong University, Ming Fan Xi'an Jiaotong University, Jian-Guang Lou Microsoft Research, Qinghua Zheng Xi'an Jiaotong University, Ting Liu Xi'an Jiaotong University | ||
00:52 6mTalk | CHASE: Checklist to Assess User Experience in Internet of Things EnvironmentsNIER New Ideas and Emerging Results Rodrigo Almeida Federal University of Ceará, Joseane Paiva Federal University of Ceará, Rossana Andrade Federal University of Ceará, Ticianne Darin Federal University of Ceará | ||