Xuewan He
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xuewan He has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Guoming Lu
University of Electronic Science and Technology of China, Chengdu, China
4 shared publications - Jielei Wang
University of Electronic Science and Technology of China
4 shared publications - Tao He
University of Electronic Science and Technology of China, Chengdu, China
2 shared publications - Zihan Cheng
University of Electronic Science and Technology of China, Chengdu, China
2 shared publications - Dongnan Liu
School of Computer Science and Engineering, University of Electronic Science and Technology of China (UESTC), Chengdu, China
1 shared publication - Guoqing Wang
University of Electronic Science and Technology of China, Chengdu, China
1 shared publication
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1177-1186
SPARK: Learning Task-Specific Semantic from Vision-Language Models for Data-Scarce Knowledge Distillation
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1177-1186
Open AccessSPARK: Learning Task-Specific Semantic from Vision-Language Models for Data-Scarce Knowledge Distillation
- 14 July 2026
- Published by Springer Nature
- p. 327-340
Dual-Domain Imitation with Normalization: Enhancing Feature Distillation for Object Detection
- 14 July 2026
- Published by Springer Nature
- p. 327-340
Dual-Domain Imitation with Normalization: Enhancing Feature Distillation for Object Detection
- Open Access
- 15 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1563-1567
CPD: Distilling Semantics in Feature Space via Class-Projection Interaction
- 15 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1563-1567
Open AccessCPD: Distilling Semantics in Feature Space via Class-Projection Interaction
- 9 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (6) , 9023-9041
Monotonic Rank Knowledge Distillation via Kendall Correlation
The computational and memory demands of deep neural networks for vision tasks remain a critical barrier to their deployment on resource-constrained edge devices. Although knowledge distillation (KD) effectively transfers over-parameterized models’ knowledge into compact students, its efficacy diminishes substantially when a significant capacity gap exists between them. Current approaches often impose linear mapping constraints between output distributions, an assumption that becomes prohibitively restrictive under such capacity gaps. This paper proposes a fundamental relaxation of alignment requirements. Specifically, rather than enforcing strict parametric relationships, we experimentally validate that preserving monotonic rank correlation between teacher and student outputs suffices for effective knowledge transfer. To operationalize this insight, we introduce Monotonic Rank Knowledge Distillation, a novel framework that leverages differentiable approximations of Kendall’s rank correlation coefficient to measure and optimize rank-order consistency. Our methodology further decomposes rank correlation into inter-class and intra-class components, ensuring the student network retains both global discriminative patterns and fine-grained categorical distinctions inherent to the teacher’s outputs. Extensive experiments across CIFAR-100 and ImageNet-1K benchmarks validate the effectiveness of our approach, demonstrating consistent performance gains over state-of-the-art distillation methods. The proposed framework achieves superior generalization across diverse architectures, including CNN-based, MLP-based, and ViT-based, with particular efficacy in various compression scenarios.
- 9 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (6) , 9023-9041
Monotonic Rank Knowledge Distillation via Kendall Correlation
The computational and memory demands of deep neural networks for vision tasks remain a critical barrier to their deployment on resource-constrained edge devices. Although knowledge distillation (KD) effectively transfers over-parameterized models’ knowledge into compact students, its efficacy diminishes substantially when a significant capacity gap exists between them. Current approaches often impose linear mapping constraints between output distributions, an assumption that becomes prohibitively restrictive under such capacity gaps. This paper proposes a fundamental relaxation of alignment requirements. Specifically, rather than enforcing strict parametric relationships, we experimentally validate that preserving monotonic rank correlation between teacher and student outputs suffices for effective knowledge transfer. To operationalize this insight, we introduce Monotonic Rank Knowledge Distillation, a novel framework that leverages differentiable approximations of Kendall’s rank correlation coefficient to measure and optimize rank-order consistency. Our methodology further decomposes rank correlation into inter-class and intra-class components, ensuring the student network retains both global discriminative patterns and fine-grained categorical distinctions inherent to the teacher’s outputs. Extensive experiments across CIFAR-100 and ImageNet-1K benchmarks validate the effectiveness of our approach, demonstrating consistent performance gains over state-of-the-art distillation methods. The proposed framework achieves superior generalization across diverse architectures, including CNN-based, MLP-based, and ViT-based, with particular efficacy in various compression scenarios.