Publications

• Can LLMs Learn by Teaching? A Preliminary Study Xuefei Ning*‡, Zifu Wang*, Shiyao Li*, Zinan Lin*‡, Peiran Yao*, Tianyu Fu, Matthew B Blaschko, Guohao Dai, Huazhong Yang, Yu Wang‡ [NeurIPS 2024] [arXiv] [code] [blog] (*Equal contribution) (‡Corresponding authors)
• Improving the Training of Rectified Flows Sangyun Lee, Zinan Lin, Giulia Fanti [NeurIPS 2024] [arXiv] [code]
• Linear Combination of Saved Checkpoints Makes Consistency and Diffusion Models Better Enshu Liu*, Junyi Zhu*, Zinan Lin+‡, Xuefei Ning+‡, Matthew B. Blaschko, Sergey Yekhanin, Shengen Yan, Guohao Dai, Huazhong Yang, Yu Wang‡ [arXiv] [code] (*Equal contribution) (+Co-advise) (‡Corresponding authors)
• Efficient Expert Pruning for Sparse Mixture-of-Experts Language Models: Enhancing Performance and Reducing Inference Costs Enshu Liu*, Junyi Zhu*, Zinan Lin+‡, Xuefei Ning+‡, Matthew B. Blaschko, Shengen Yan, Guohao Dai, Huazhong Yang, Yu Wang‡ [arXiv] [code] (*Equal contribution) (+Co-advise) (‡Corresponding authors)
• Skeleton-of-Thought: Prompting LLMs for Efficient Parallel Generation [Previous Title] Skeleton-of-Thought: Large Language Models Can Do Parallel Decoding Xuefei Ning*, Zinan Lin*, Zixuan Zhou*, Zifu Wang, Huazhong Yang, Yu Wang [ICLR 2024] [NeurIPS 2023 Workshop on Efficient Natural Language and Speech Processing] [arXiv] [website] [blog] [code] (*Equal contribution)
• Differentially Private Synthetic Data via Foundation Model APIs 1: Images Zinan Lin, Sivakanth Gopi, Janardhan Kulkarni, Harsha Nori, Sergey Yekhanin (NeurIPS Workshop Oral) [ICLR 2024] [NeurIPS 2023 Workshop on Synthetic Data Generation with Generative AI] [arXiv] [code] [blog] [invited talk at AI Time (in Chinese)]
• Differentially Private Synthetic Data via Foundation Model APIs 2: Text Chulin Xie, Zinan Lin, Arturs Backurs, Sivakanth Gopi, Da Yu, Huseyin Inan, Harsha Nori, Haotian Jiang, Huishuai Zhang, Yin Tat Lee, Bo Li, Sergey Yekhanin (ICML Spotlight) [ICML 2024] [ICLR 2024 Workshop on Secure and Trustworthy Large Language Models] [arXiv] [code] [website] [blog]
• Statistic Maximal Leakage Shuaiqi Wang, Zinan Lin, Giulia Fanti [ISIT 2024]
• Privacy-Preserving In-Context Learning with Differentially Private Few-Shot Generation Xinyu Tang, Richard Shin, Huseyin A. Inan, Andre Manoel, Fatemehsadat Mireshghallah, Zinan Lin, Sivakanth Gopi, Janardhan Kulkarni, Robert Sim [ICLR 2024] [arXiv] [code] [blog]
• Efficiently Computing Similarities to Private Datasets Arturs Backurs, Zinan Lin, Sepideh Mahabadi, Sandeep Silwal, Jakub Tarnawski [ICLR 2024] [arXiv]
• Rescaling Intermediate Features Makes Trained Consistency Models Perform Better Junyi Zhu, Zinan Lin, Enshu Liu, Xuefei Ning, Matthew B. Blaschko [ICLR 2024 (TinyPapers)]
• Summary Statistic Privacy in Data Sharing Zinan Lin*, Shuaiqi Wang*, Vyas Sekar, Giulia Fanti [JSAIT 2024] [arXiv] (*Equal contribution)
• Mixture-of-Linear-Experts for Long-term Time Series Forecasting Ronghao Ni, Zinan Lin, Shuaiqi Wang, Giulia Fanti [AISTATS 2024] [arXiv] [code] [blog]
• FlashEval: Towards Fast and Accurate Evaluation of Text-to-image Diffusion Generative Models Lin Zhao*, Tianchen Zhao*, Zinan Lin, Xuefei Ning, Guohao Dai, Huazhong Yang, Yu Wang [CVPR 2024] [arXiv] (*Equal contribution)
• Selective Pre-training for Private Fine-tuning Da Yu, Sivakanth Gopi, Janardhan Kulkarni, Zinan Lin, Saurabh Naik, Tomasz Lukasz Religa, Jian Yin, Huishuai Zhang [TMLR] [arXiv]
• MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization Tianchen Zhao*, Xuefei Ning*, Tongcheng Fang*, Enshu Liu, Guyue Huang, Zinan Lin, Shengen Yan, Guohao Dai, Yu Wang [ECCV 2024] [arXiv] [code] [Hugging Face Pipeline] [website] (*Equal contribution)
• ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation Tianchen Zhao, Tongcheng Fang, Enshu Liu, Wan Rui, Widyadewi Soedarmadji, Shiyao Li, Zinan Lin, Guohao Dai, Shengen Yan, Huazhong Yang, Xuefei Ning, Yu Wang [arXiv] [code] [website]

• OMS-DPM: Optimizing the Model Schedule for Diffusion Probabilistic Models Enshu Liu*, Xuefei Ning*, Zinan Lin*, Huazhong Yang, Yu Wang [ICML 2023] [arXiv] [code] [website] (*Equal contribution)
• DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models Boxin Wang*, Weixin Chen*, Hengzhi Pei*, Chulin Xie*, Mintong Kang*, Chenhui Zhang*, Chejian Xu, Zidi Xiong, Ritik Dutta, Rylan Schaeffer, Sang T Truong, Simran Arora, Mantas Mazeika, Dan Hendrycks, Zinan Lin, Yu Cheng, Sanmi Koyejo, Dawn Song, Bo Li* (NeurIPS Outstanding Paper & Oral) [NeurIPS 2023] [arXiv] [website] [blog] [code] (*Lead authors)
• Training Private and Efficient Language Models with Synthetic Data from LLMs Da Yu, Arturs Backurs, Sivakanth Gopi, Huseyin Inan, Janardhan Kulkarni, Zinan Lin, Chulin Xie, Huishuai Zhang, Wanrong Zhang [NeurIPS 2023 Workshop on Socially Responsible Language Modelling Research]

• Distributional Privacy for Data Sharing Zinan Lin*, Shuaiqi Wang*, Vyas Sekar, Giulia Fanti [NeurIPS 2022 Workshop SyntheticData4ML] (*Equal contribution)
• Practical GAN-based Synthetic IP Header Trace Generation using NetShare Yucheng Yin, Zinan Lin, Minhao Jin, Giulia Fanti, Vyas Sekar [SIGCOMM 2022] [code] [talk]
• RareGAN: Generating Samples for Rare Classes Zinan Lin, Hao Liang, Giulia Fanti, Vyas Sekar [AAAI 2022] [arXiv] [code] [talk & slide] [invited blog at AI Time (in Chinese)] [invited talk at AI Time (in Chinese)]
• PcapShare: Exploring the Feasibility of GANs for Synthetic Packet Header Trace Generation Yucheng Yin, Zinan Lin, Minhao Jin, Giulia Fanti, Vyas Sekar [COMSNETS 2022 (demo)]

• Why Spectral Normalization Stabilizes GANs: Analysis and Improvements Zinan Lin, Vyas Sekar, Giulia Fanti [NeurIPS 2021] [arXiv] [code] [talk & slide] [invited talk at AI Time (in Chinese)] [blog]
• On the Privacy Properties of GAN-generated Samples Zinan Lin, Vyas Sekar, Giulia Fanti [AISTATS 2021] [arXiv] [talk & slide]
• Pareto GAN: Extending the Representational Power of GANs to Heavy-Tailed Distributions Todd Huster, Jeremy E.J. Cohen, Zinan Lin, Kevin Chan, Charles Kamhoua, Nandi Leslie, Cho-Yu Jason Chiang, Vyas Sekar [ICML 2021] [arXiv]
• MLGO: a Machine Learning Guided Compiler Optimizations Framework Mircea Trofin*, Yundi Qian*, Eugene Brevdo, Zinan Lin, Krzysztof Choromanski, David Li [arXiv] [code] (*Equal contribution)

• Using GANs for Sharing Networked Timeseries Data: Challenges, Initial Promise, and Open Questions [Previous Title] Generating High-fidelity, Synthetic Time Series Datasets with DoppelGANger Zinan Lin, Alankar Jain, Chen Wang, Giulia Fanti, Vyas Sekar (IMC Best Paper Finalist) [IMC 2020] [arXiv] [code] [talk] [blog]
• InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANs [Previous Title] InfoGAN-CR: Disentangling Generative Adversarial Networks with Contrastive Regularizers Zinan Lin, Kiran Thekumparampil, Giulia Fanti, Sewoong Oh [ICML 2020] [arXiv] [code] [talk & slide]

• Towards Oblivious Network Analysis using Generative Adversarial Networks Zinan Lin, Soo-Jin Moon, Carolina M. Zarate, Ritika Mulagalapalli, Sekar Kulandaivel, Giulia Fanti, Vyas Sekar [HotNets 2019] [talk]

• PacGAN: The Power of Two Samples in Generative Adversarial Networks Zinan Lin, Ashish Khetan, Giulia Fanti, Sewoong Oh [NeurIPS 2018] [JSAIT 2020] [arXiv] [poster] [website] [interview] [code] [invited talk at yanxishe.com (in Chinese)]
• Robustness of Conditional GANs to Noisy Labels Kiran Thekumparampil, Ashish Khetan, Zinan Lin, Sewoong Oh (NeurIPS Spotlight) [NeurIPS 2018] [arXiv] [code]
• RNN-SM: Fast Steganalysis of VoIP Streams Using Recurrent Neural Network Zinan Lin, Yongfeng Huang, Jilong Wang [IEEE Transactions on Information Forensics and Security] [PDF (3.5MB)] [code]