Publication

January 2023 -- One paper is accepted to ISSTA 2023.
Our paper, DyCL, is accepted to ISSTA 2023. DyCL is a automatic tool that rewrites the dynamic neural networks for supporting existing DL compilers.
Jan 16, 2023 3 min read Publication
April 2022 -- One paper is accepted to ASE 2022.
Our paper, DeepPerform, is accepted to ASE 2022. DeepPerform is designed to test the performance robustness of adaptive neural networks.
Aug 30, 2022 3 min read Publication
June 2022 -- One paper is accepted to ESEC/FSE 2022.
Our paper, NMTSloth, is accepted to ESEC/FSE 2022. NMTSloth is designed to test the performance degradation bugs in neural machine translation systems. We show NMTSloth can generate generate inputs that increase neural machine translation systems computational consumption upto 3153%.
Jun 14, 2022 3 min read Publication
April 2022 -- One paper is accepted to IJCAI 2022.
Our paper, NNReverse, is accepted to IJCAI 2022. NNReverse is an automatic approach for reversing DNNs from AI programs.
Apr 22, 2022 3 min read Publication
April 2022 -- Serve as a student volunteer in ICSE 2022
The template is mobile first with a responsive design to ensure that your site looks stunning on every device. ## Get Started - 👉 [**Create a new site**](https://wowchemy.
Apr 6, 2022 3 min read Publication
Program Committee Member -- MSR2023, CVPR 2023, AAAI 2023, WACV 2022, ECCV 2022
The template is mobile first with a responsive design to ensure that your site looks stunning on every device. ## Get Started - 👉 [**Create a new site**](https://wowchemy.
Apr 6, 2022 Publication
Sub-Reviewer -- ICSE 2023, ICST 2023, ASE 2022, ICST 2022, ICSE 2021
The template is mobile first with a responsive design to ensure that your site looks stunning on every device. ## Get Started - 👉 [**Create a new site**](https://wowchemy.
Apr 6, 2022 Publication
March 2022 -- One paper is accepted to CVPR 2022.
Our paper, NICGSlowDown, is accepted to CVPR 2022. To the best of our knowledge, NICGSlowDown is the first work to evaluate the efficiency robustness of neural caption generation models.
Mar 3, 2022 3 min read Publication
November 2020 -- Serve as an additional reviewers in ICSE 2021 technical track.
I will serve as a Committee Member in Additional Reviewers within the Technical Track-track in ICSE 2021
Nov 10, 2020 3 min read Publication
May 2020 -- One paper is accepted to ESEC/FSE 2020.
Our paper, DENAS, is accepted to ESEC/FSE 2020. DENAS is designed to extract human-understandable rules from well-trained DNNs. Furthermore, DENAS can help model developers locate the DNN bugs and fix the DNN bugs through targeted retraining.
May 10, 2020 3 min read Publication