研究
我主要研究生成模型加速、神经形态计算与高效模型。研究兴趣涵盖脑启发计算(脉冲神经网络与神经形态系统)、高效建模(加速算法、剪枝与量化),以及 AI 安全(对抗攻击与模型可解释性)。
博士期间主要围绕脉冲神经网络、神经形态计算与高效模型开展研究。同时,我也在阿里巴巴 Wan 团队实习,探索视频生成模型的加速方法。
Google Scholar已发表工作
2026
-
Proxy target: Bridging the gap between discrete spiking neural networks and continuous control
Advances in Neural Information Processing Systems 38, 159158-159184会议论文
-
Uncertainty-Aware Token Importance Estimation in Spiking Transformers
arXiv preprint arXiv:2605.09276预印本
-
Activity Pruning for Efficient Spiking Neural Networks
Advances in Neural Information Processing Systems 38, 171134-171153会议论文
-
Generalized Threshold Optimization with Harmony Multi-Threshold Neurons for Accurate ANN-to-SNN Conversion
Proceedings of the AAAI Conference on Artificial Intelligence 40 (3), 2227-2235会议论文
-
Reducing ANN-SNN Conversion Error via Residual Membrane Potential Alignment
arXiv preprint arXiv:2608.13952预印本
-
UniSparse: Combining Weight Pruning and Spike Sparsification in Spiking Neural Networks
The 43rd International Conference on Machine Learning会议论文
2025
-
Differential coding for training-free ANN-to-SNN conversion
arXiv preprint arXiv:2503.00301预印本
-
Ultrafast probabilistic neuron in an artificial spin ice for robust deep neural networks
Advanced Functional Materials 35 (11), 2417334期刊论文
-
PredNext: Explicit Cross-View Temporal Prediction for Unsupervised Learning in Spiking Neural Networks
arXiv preprint arXiv:2509.24844预印本
2024
-
Towards high-performance spiking transformers from ANN to SNN conversion
Proceedings of the 32nd ACM International Conference on Multimedia, 10688-10697会议论文
-
A progressive training framework for spiking neural networks with learnable multi-hierarchical model
International Conference on Learning Representations 2024, 4925-4941会议论文
-
Enhancing adversarial robustness in SNNs with sparse gradients
arXiv preprint arXiv:2405.20355预印本
-
Threaten spiking neural networks through combining rate and temporal information
International Conference on Learning Representations 2024, 55658-55674会议论文
-
Robust decoding of rich dynamical visual scenes with retinal spikes
IEEE Transactions on Neural Networks and Learning Systems 36 (2), 3396-3409期刊论文
-
Inference-Scale Complexity in ANN-SNN Conversion for High-Performance and Low-Power Applications
arXiv preprint arXiv:2409.03368预印本
2023
-
Optimal ANN-SNN conversion for high-accuracy and ultra-low-latency spiking neural networks
arXiv preprint arXiv:2303.04347预印本
-
Reducing ANN-SNN conversion error through residual membrane potential
Proceedings of the AAAI Conference on Artificial Intelligence 37 (1), 11-21会议论文
-
Bridging the gap between ANNs and SNNs by calibrating offset spikes
arXiv preprint arXiv:2302.10685预印本
-
Rate gradient approximation attack threats deep spiking neural networks
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition会议论文
-
Red teaming deep neural networks with feature synthesis tools
Advances in Neural Information Processing Systems 36, 80470-80516会议论文
2022
-
Optimized potential initialization for low-latency spiking neural networks
Proceedings of the AAAI Conference on Artificial Intelligence 36 (1), 11-20会议论文
-
SNN-RAT: Robustness-enhanced spiking neural network through regularized adversarial training
Advances in Neural Information Processing Systems 35, 24780-24793会议论文
-
Decoding pixel-level image features from two-photon calcium signals of macaque visual cortex
Neural Computation 34 (6), 1369-1397期刊论文