Scalability and Precision Improvement of Neural Program Synthesis
Mosts of the neural synthesis construct encoder-decoder models to learn a probability distribution over the space of programs. Two drawbacks in such neural program synthesis are that the synthesis scale is relatively small and the correctness of the synthesis result cannot be guaranteed.We address these problems by constructing a framework, which analyzes and solves problems from three dimensions: program space description, model architecture, and result processing. Experiments show that the scalability and precision of synthesis are improved in every dimension.
Wed 23 Sep Times are displayed in time zone: (UTC) Coordinated Universal Time
02:20 - 03:20: SRC Poster (1)Student Research Competition at Koala Chair(s): Chunyang ChenMonash University | |||
02:20 - 02:25 Poster | Dynamic Algorithm Selection for SMT Student Research Competition Nikhil PimpalkhareUniversity of California, Berkeley | ||
02:25 - 02:30 Poster | CrossPriv: User Privacy Preservation Model for Cross-Silo Federated Software Student Research Competition Harshita DiddeeGuru Gobind Singh Indraprastha University, Bhrigu KansraGuru Gobind Singh Indraprastha University | ||
02:30 - 02:35 Poster | Scalability and Precision Improvement of Neural Program Synthesis Student Research Competition Yating ZhangNational University of Defense Technology | ||
02:35 - 02:40 Poster | The Classification and Propagation of Program CommentsSecond place SRC - Undergraduate Student Research Competition Xiangzhe XuNanjing University | ||
02:40 - 02:45 Poster | Source Code and Binary Level Vulnerability Detection and Hot PatchingThird place SRC - Graduate Student Research Competition Zhengzi XuNanyang Technological University | ||
02:45 - 02:50 Poster | FLUX: From SQL to GQL query translation tool Student Research Competition Chandan SharmaAuckland University of Technology | ||
02:50 - 02:55 Poster | Discovering UI Display Issues with Visual UnderstandingWinner SRC - Graduate Student Research Competition Zhe LiuLaboratory for Internet Software Technologies, Institute of Software Chinese Academy of Sciences, University of Chinese Academy of Sciences |