Publications
Publications by categories in reversed chronological order. * indicates co-first author or co-corresponding author.
2024
- DISP-LLM: Dimension-Independent Structural Pruning for Large Language ModelsIn Advances in Neural Information Processing Systems (NeurIPS) (to appear), 2024
- Unlocking Memorization in Large Language Models with Dynamic Soft PromptingIn Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP) (to appear), 2024
- NAACLAdaptive Rank Selections for Low-Rank Approximation of Language ModelsIn Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), 2024
- Auto-Train-Once: Controller Network Guided Automatic Network Pruning from ScratchIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
- Device-Wise Federated Network PruningIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
- Jointly Training and Pruning CNNs via Learnable Agent Guidance and AlignmentIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
- BilevelPruning: Unified Dynamic and Static Channel Pruning for Convolutional Neural NetworksIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
- Compressing Image-to-Image Translation GANs Using Local Density Structures on Their Learned ManifoldIn Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2024
- WACVToken Fusion: Bridging the Gap between Token Pruning and Token MergingIn Winter Conference on Computer Vision (WACV), 2024
2023
- Learning to Jointly Share and Prune Weights for Grounding Based Vision and Language ModelsIn International Conference on Learning Representations (ICLR), 2023
- EffConv: Efficient Learning of Kernel Sizes for Convolution Layers of CNNsIn Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2023
- Structural alignment for network pruning through partial regularizationProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2023
- TPAMIGradient Descent Ascent for Minimax Problems on Riemannian ManifoldsIEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2023
- Dynamic Low-rank Estimation for Transformer-based Language ModelsIn Proceedings of the Conference on Empirical Methods in Natural Language Processing Findings (EMNLP findings), 2023
2022
- Bregman Gradient Policy OptimizationIn International Conference on Learning Representations (ICLR), 2022
- Disentangled differentiable network pruningIn European Conference on Computer Vision (ECCV), 2022
- Interpretations steered network pruning via amortized inferred saliency mapsIn European Conference on Computer Vision (ECCV), 2022
- Recover fair deep classification models via altering pre-trained structureIn European Conference on Computer Vision (ECCV), 2022
- Riemannian gradient methods for stochastic composition problemsNeural Networks, 2022
- JMLRAccelerated zeroth-order and first-order momentum methods from mini to minimax optimizationThe Journal of Machine Learning Research (JMLR), 2022
- Improving social network embedding via new second-order continuous graph neural networksIn Proceedings of the 28th ACM SIGKDD conference on knowledge discovery and data mining (KDD), 2022
- Enhanced bilevel optimization via bregman distanceAdvances in Neural Information Processing Systems, 2022
2021
- Network pruning via performance maximizationIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021
- Exploration and estimation for model compressionIn Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021
- Adversarial attack on deep cross-modal hamming retrievalIn Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021
- JMLRBlack-box reductions for zeroth-order gradient algorithms to achieve lower query complexityThe Journal of Machine Learning Research (JMLR), 2021
2020
- Discrete model compression with resource constraint for deep neural networksIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020
- Momentum-based policy gradient methodsIn International conference on machine learning (ICML), 2020
2019
- Cross Domain Model Compression by Structurally Weight SharingIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2019
- Cross-modal learning with adversarial samplesAdvances in neural information processing systems (NeurIPS), Jun 2019
- Zeroth-order stochastic alternating direction method of multipliers for nonconvex nonsmooth optimizationInternational Joint Conferences on Artificial Intelligence (IJCAI), Jun 2019
2018
- Action prediction from videos via memorizing hard-to-predict samplesIn Proceedings of the AAAI conference on artificial intelligence (AAAI), Jun 2018
2017
- Video recovery via learning variation and consistency of imagesIn Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), Jun 2017
2016
- Discriminative multi-instance multitask learning for 3D action recognitionIEEE Transactions on Multimedia, Jun 2016