Yuantian Ding

I’m Yuantian Ding, a 4th-year PhD student at Purdue University specializing in program synthesis and verification under the supervision of Prof. Xiaokang Qiu and Prof. Jingbo Wang. I’m currently working on neural network verification.

Education

Purdue University (2022 - Present)
Ph.D. Electrical and Computer Engineering
Advised by Xiaokang Qiu

University of Science and Technology of China (2018 - 2022)
B.E. in Computer Science and Technology
GPA: 3.97/4.3 (Ranked 6/251)

Projects

BoundLab: Neural Network Verification using Sparse Polynomial.
A VNN tool to precisely verify transformers with additional cost.
SITE

DryadSynth: Dryad Synthesizer for SyGuS competition
A SyGuS solver designed by Purdue CAP, under active development.
SITE CODE

Publication

A Concurrent Approach to String Transformation Synthesis.
Yuantian Ding, Xiaokang Qiu.
(PLDI ’25) Proc. 46th ACM SIGPLAN Conference on Programming Language Design and Implementation.
PDF DOI CODE TALK SLIDES

How Many Quantum Circuit Identities Are Needed to Generate All Others?
Yuantian Ding, Nengkun Yu, Xiaokang Qiu.
(CAV ’26) Proc. 38th International Conference on Computer Aided Verification PDF CODE

DiVerT: Differential Verification of Transformers
Paul Jiang*, Yuantian Ding*, Saurabh Bagchi, Ðorde Žikelic, Jingbo Wang.
Manualscript submitted to ICSE 27

Experience

Nanjing University Research Internship (07/2021 - 12/2021)
with Zhiqiang Zuo, Harry Xu
Topic: A New Serverless Platform with Customized JIT Design

SRI SSFT24: The 13th Summer School on Formal Techniques (07/2021 - 12/2021)
Summer school for learning techniques based on formal logic.

Awards

ASC Student Supercomputer Challenge 2021 First Prize
Topic (My Part): Efficient Parallelization of PRESTO

ACM-China International Parallel Computing Challenge 2020 Third Prize
Topic (Final): Efficient Numerical Simulation of Supersonic Underexpanded Thermal Jets