Projects per year
Personal profile
Personal profile
Dr. Pascal Lefèvre is an Assistant Professor in the School of AI and Advanced Computing. He received a Master's Degree in Cryptography and Computer Security from Bordeaux University, France, and then his Ph.D. from Poitiers University and XLIM Laboratory, France. He joined Shenzhen University as a Post-Doctoral researcher and later, worked in AI and Computer Vision in the tech industry. His past research involved multiple fields such as Cryptography, Error-Correcting Codes, Signal and Image Processing, Multimedia Security, and Artificial Intelligence. He actively works on Neural Network Compression and the applications of AI for Mathematics.
Research Development Funding: "Research on a new paradigm for neural network compression". Funding: 100,000 RMB.
Funded External Research Grant for Industry Applications: "Fast-AI: Neural Network Compression Methods". Funding amount: 2,630,000 RMB.
Open PhD position: Seeking a serious, highly motivated PhD student to tackle cutting-edge challenges in Neural Network Compression. Your work will directly shape the development of efficient AI for real-world systems. Requires strong foundations in deep learning and relentless drive. Please send your full resume to Pascal.Lefevre@xjtlu.edu.cn.
Research interests
Neural Network Compression
Linear Algebra / Arithmetic
Automated Theorem Proving / Lean 4 / AI for Mathematics
Computer Vision / Machine Learning / Deep Learning
Multimedia Security / Error-Correcting Codes / Cryptography
Experience
Assistant Professor, School of AI and Advanced Computing, Xi’an Jiaotong-Liverpool University, 2023 - now
AI R&D, Shenzhen Tech Industry, 2021 - 2023
Post-Doctoral Researcher, Shenzhen University, 2019 - 2021
Teaching
Signal and Image Processing
Mathematics for AI and Data Science
Design of Algorithms and Analysis
Awards and honours
Best student paper award, CORESA 2017, France
Best oral presentation, ICIVC 2025, Chengdu, China
Related documents
Education/Academic qualification
Ph.D., Signal and Image Processing, Poitiers University and XLIM Laboratory, France, 2018
Msc, Cryptology and Computer Security, Bordeaux University, 2014
BSc, Pure Mathematics and Computer Science, French Polynesia University, 2012
Research areas
- Neural Network Compression
- Deep Learning
- Automated Theorem Proving
- Multimedia Security
- Linear Algebra & Optimization
Person Types
- Staff
Fingerprint
- 1 Similar Profiles
Collaborations and top research areas from the last five years
Projects
- 2 Active
-
Fast-IA: Méthode de compression de réseaux de neurones
LEFEVRE, P. (CoPI)
24/11/25 → …
Project: Collaborative Research Project
-
Research on a new paradigm for neural network compression
LEFEVRE, P. (PI)
1/01/25 → 31/12/27
Project: Internal Research Project
-
A Review of Neural Network Compression Baseline Algorithms and Comparative Study Analysis
Fu, W., Strauss, T., Zhou, M. & LEFEVRE, P., 2026, A Review of Neural Network Compression Baseline Algorithms and Comparative Study Analysis. 2026 11th International Conference on Signal and Image Processing (ICSIP 2026)Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review
Open AccessFile98 Downloads (Pure) -
Procédés de compression de données numériques, de décompression, de traitement et dispositifs correspondants
LEFEVRE, P. (Inventor) & Gaborit, P. (Inventor), 21 May 2026Translated title of the contribution :Digital data compression, decompression, processing methods and corresponding devices Research output: Patent
-
Teaching Mathematics in the Age of AI: A Teaching Framework for Undergraduate-Level Combinatorics
LEFEVRE, P. & 炯龙苏, 2026, Teaching Mathematics in the Age of AI: A Teaching Framework for Undergraduate-Level Combinatorics. ICAIE 2026, 2026 2nd International Conference on Artificial Intelligence and EducationResearch output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review
Open AccessFile142 Downloads (Pure) -
A Reversible Grayscale Method Based on Bit-Field Multi-Channel Fusion Encoding
Wang, T., Pan, W., Yang, Y. & LEFEVRE, P., 23 Jun 2025, A Reversible Grayscale Method Based on Bit-Field Multi-Channel Fusion Encoding. Chengdu, China: Institute of Electrical and Electronics Engineers Inc., p. 555 560 p.Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review
Open Access -
Halfway Escape Optimization: A Quantum-Inspired Solution for Complex Optimization Problems
Li, J., PP Majeed, A. & LEFEVRE, P., 1 Sept 2024, (Submitted) In: Evolving Systems.Research output: Contribution to journal › Article › peer-review
Activities
-
2026 2nd International Conference on Artificial Intelligence and Education
LEFEVRE, P. (Speaker)
13 May 2026 → 15 May 2026Activity: Talk or presentation › Presentation at conference/workshop/seminar
-
2025 Conference on Multi-Agent Applications in China
LEFEVRE, P. (Participant)
20 Dec 2025Activity: Participating in or organising an event › Participating in an event e.g. a conference, workshop, …
-
On the Role of AI in Scientific Discovery - From Signal Signatures to Collaborative Mathematical Discovery
LEFEVRE, P. (Speaker)
22 Aug 2025Activity: Talk or presentation › Invited talk
-
Workshop on Time-Frequency Analysis and Its Applications
LEFEVRE, P. (Participant)
22 Aug 2025Activity: Participating in or organising an event › Participating in an event e.g. a conference, workshop, …
-
8th International Conference on Big Data and Artificial Intelligence (BDAI 2025)
LEFEVRE, P. (Participant)
22 Aug 2025 → 24 Aug 2025Activity: Participating in or organising an event › Participating in an event e.g. a conference, workshop, …