no code implementations • 25 Apr 2024 • Zhensu Sun, Xiaoning Du, Zhou Yang, Li Li, David Lo
Particularly, abundant grammar tokens and formatting tokens are included to make the code more readable to humans.
1 code implementation • 21 Mar 2024 • Mingze Ni, Zhensu Sun, Wei Liu
Recent studies on adversarial examples expose vulnerabilities of natural language processing (NLP) models.
1 code implementation • 18 Jan 2024 • Zhensu Sun, Xiaoning Du, Fu Song, Shangwen Wang, Li Li
These findings motivate our exploration of dynamic inference in code completion and inspire us to enhance it with a decision-making mechanism that stops the generation of incorrect code.
1 code implementation • 28 Aug 2023 • Zhensu Sun, Xiaoning Du, Fu Song, Li Li
Even worse, the ``black-box'' nature of neural models sets a high barrier for externals to audit their training datasets, which further connives these unauthorized usages.
1 code implementation • 31 May 2023 • Terry Yue Zhuo, Zhou Yang, Zhensu Sun, YuFei Wang, Li Li, Xiaoning Du, Zhenchang Xing, David Lo
This paper fills this gap by conducting a comprehensive and integrative survey of data augmentation for source code, wherein we systematically compile and encapsulate existing literature to provide a comprehensive overview of the field.
1 code implementation • 1 Mar 2023 • Mingze Ni, Zhensu Sun, Wei Liu
In response, this study proposes a new method called the Fraud's Bargain Attack (FBA), which uses a randomization mechanism to expand the search space and produce high-quality adversarial examples with a higher probability of success.
no code implementations • 13 Sep 2022 • Zhensu Sun, Xiaoning Du, Fu Song, Shangwen Wang, Mingze Ni, Li Li
The experimental results show that the proposed estimator helps save 23. 3% of computational cost measured in floating-point operations for the code completion systems, and 80. 2% of rejected prompts lead to unhelpful completion
1 code implementation • 14 Feb 2022 • Zhensu Sun, Yan Liu, Xiaoning Du, Li Li
The performance of neural code search is significantly influenced by the quality of the training data from which the neural models are derived.
1 code implementation • 25 Oct 2021 • Zhensu Sun, Xiaoning Du, Fu Song, Mingze Ni, Li Li
Github Copilot, trained on billions of lines of public code, has recently become the buzzword in the computer science research and practice community.
no code implementations • 24 May 2020 • Zhensu Sun, Yan Liu, Ziming Cheng, Chen Yang, Pengyu Che
In this work, we would like to make recommendations based on requirement descriptions to avoid these problems.