汪子乔｜同济大学｜学者主页

期刊论文

2026

A word-context classification view of noise distributions in SGN-like representation learning

De-Zhi Liu, Richong Zhang, Ziqiao Wang

Neurocomputing

DOI

2026

Fundamental Trade-Offs in Multi-Bit Watermarking of Stochastic Processes

Haiyun He, Yepeng Liu, Zhuo-Er Shen, Ziqiao Wang, Yongyi Mao, Yuheng Bu

arXiv.org

DOI

2026

On the Blessing of Pre-training in Weak-to-Strong Generalization

Wei Yao, Zhaoyang Wang, Geng-Ze Xu, Chen Qian, Dong-Rui Liu, Ziqiao Wang, Yong Liu, Yunbei Xu

arXiv.org

DOI

2025

Generalization in Federated Learning: A Conditional Mutual Information Framework

Ziqiao Wang, Cheng Long, Yongyi Mao

International Conference on Machine Learning

DOI

2025

Preserving Label Correlation for Multi-label Text Classification by Prototypical Regularizations

Fan-Shuang Kong, Richong Zhang, Xiaohui Guo, J. Chen, Ziqiao Wang

The Web Conference

DOI

2025

Distributional Information Embedding: A Framework for Multi-Bit Watermarking

Haiyun He, Yepeng Liu, Ziqiao Wang, Yongyi Mao, Yuheng Bu

2025 Asia Pacific Workshop on Data Science and Information Theory (APWDSIT)

DOI

2025

Revisiting Weak-to-Strong Generalization in Theory and Practice: Reverse KL vs. Forward KL

Wei Yao, Wenkai Yang, Ziqiao Wang, Yankai Lin, Yong Liu

DOI

2025

Dynamic Task Vector Grouping for Efficient Multi-Task Prompt Tuning

Peiyi Zhang, Richong Zhang, Zhijie Nie, Ziqiao Wang

DOI

2024

Cross-Modal and Uni-Modal Soft-Label Alignment for Image-Text Retrieval

Hailang Huang, Zhijie Nie, Ziqiao Wang, Ziyu Shang

AAAI Conference on Artificial Intelligence

DOI

2024

On Unsupervised Domain Adaptation: Pseudo Label Guided Mixup for Adversarial Prompt Tuning

Fan-Shuang Kong, Richong Zhang, Ziqiao Wang, Yongyi Mao

AAAI Conference on Artificial Intelligence

DOI

2024

LH-Mix: Local Hierarchy Correlation Guided Mixup over Hierarchical Prompt Tuning

Fan-Shuang Kong, Richong Zhang, Ziqiao Wang

Knowledge Discovery and Data Mining

DOI

2024

On f-Divergence Principled Domain Adaptation: An Improved Framework

Ziqiao Wang, Yongyi Mao

Neural Information Processing Systems

DOI

2024

Generalization Bounds via Conditional f-Information

Ziqiao Wang, Yongyi Mao

Neural Information Processing Systems

DOI

2024

MOMA: Masked Orthogonal Matrix Alignment for Zero-Additional-Parameter Model Merging

Fan-Shuang Kong, Richong Zhang, Zhijie Nie, Hang Zhou, Ziqiao Wang, Qian Sun, Chun-Ming Hu

2024

Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach

Haiyun He, Yepeng Liu, Ziqiao Wang, Yongyi Mao, Yuheng Bu

Neural Information Processing Systems

DOI

2024

Activated Parameter Locating via Causal Intervention for Model Merging

Fan-Shuang Kong, Richong Zhang, Ziqiao Wang

arXiv.org

DOI

2023

Tighter Information-Theoretic Generalization Bounds from Supersamples

Ziqiao Wang, Yongyi Mao

International Conference on Machine Learning

DOI

2023

Over-training with Mixup May Hurt Generalization

Zixuan Liu, Ziqiao Wang

International Conference on Learning Representations

DOI

2023

Sample-Conditioned Hypothesis Stability Sharpens Information-Theoretic Generalization Bounds

Ziqiao Wang, Yongyi Mao

Neural Information Processing Systems

DOI

2022

Two Facets of SDE Under an Information-Theoretic Lens: Generalization of SGD via Training Trajectories and via Terminal States

Ziqiao Wang, Yongyi Mao

Conference on Uncertainty in Artificial Intelligence

DOI

2022

Information-Theoretic Analysis of Unsupervised Domain Adaptation

Ziqiao Wang, Yongyi Mao

International Conference on Learning Representations

DOI

2021

On the Generalization of Models Trained with SGD: Information-Theoretic Bounds and Implications

Ziqiao Wang, Yongyi Mao

International Conference on Learning Representations

2020

On SkipGram Word Embedding Models with Negative Sampling: Unified Framework and Impact of Noise Distributions

Ziqiao Wang, Yongyi Mao, Hongyu Guo, Richong Zhang

arXiv.org

预印本

2026

On the Recoverability of Private Information Unlearning in Large Language Models

Shicheng Hu, Runzhi Tian, Ziqiao Wang, Yongyi Mao

arXiv (Cornell University)

DOI

2026

Beyond Penalization: Diffusion-based Out-of-Distribution Detection and Selective Regularization in Offline Reinforcement Learning

Qingjun Wang, Hongtu Zhou, Hang Yu, Junqiao Zhao, Yanping Zhao, Chen Ye, Ziqiao Wang, Guang Chen

arXiv (Cornell University)

DOI

2026

Learning Rules, Memorizing Facts: An Information-Theoretic View of Hallucination

Gengze Xu, Anxin Guo, Yifei Hu, Xiangyu Wu, Yong Liu, Ziqiao Wang

arXiv (Cornell University)

DOI

2025

The Capabilities and Limitations of Weak-to-Strong Generalization: Generalization and Calibration

Yao, Wei, Wenkai Yang, Xu, Gengze, Ziqiao Wang, Yankai Lin, Yong Liu

arXiv (Cornell University)

DOI

2025

On Weak-to-Strong Generalization and f-Divergence

Yao, Wei, Gengze Xu, Huayi Tang, Wenkai Yang, Donglin Di, Ziqiao Wang, Yong Liu

arXiv (Cornell University)

DOI

2025

Weak-to-Strong Generalization via Bregman Bias-Variance Decomposition

Gengze Xu, Yao, Wei, Ziqiao Wang, Yong Liu

arXiv (Cornell University)

DOI

汪子乔

精选作品

期刊论文

A word-context classification view of noise distributions in SGN-like representation learning

Fundamental Trade-Offs in Multi-Bit Watermarking of Stochastic Processes

On the Blessing of Pre-training in Weak-to-Strong Generalization

Generalization in Federated Learning: A Conditional Mutual Information Framework

Preserving Label Correlation for Multi-label Text Classification by Prototypical Regularizations

Distributional Information Embedding: A Framework for Multi-Bit Watermarking

Revisiting Weak-to-Strong Generalization in Theory and Practice: Reverse KL vs. Forward KL

Dynamic Task Vector Grouping for Efficient Multi-Task Prompt Tuning

Cross-Modal and Uni-Modal Soft-Label Alignment for Image-Text Retrieval

On Unsupervised Domain Adaptation: Pseudo Label Guided Mixup for Adversarial Prompt Tuning

LH-Mix: Local Hierarchy Correlation Guided Mixup over Hierarchical Prompt Tuning

On f-Divergence Principled Domain Adaptation: An Improved Framework

Generalization Bounds via Conditional f-Information

MOMA: Masked Orthogonal Matrix Alignment for Zero-Additional-Parameter Model Merging

Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach

Activated Parameter Locating via Causal Intervention for Model Merging

Tighter Information-Theoretic Generalization Bounds from Supersamples

Over-training with Mixup May Hurt Generalization

Sample-Conditioned Hypothesis Stability Sharpens Information-Theoretic Generalization Bounds

Two Facets of SDE Under an Information-Theoretic Lens: Generalization of SGD via Training Trajectories and via Terminal States

Information-Theoretic Analysis of Unsupervised Domain Adaptation

On the Generalization of Models Trained with SGD: Information-Theoretic Bounds and Implications

On SkipGram Word Embedding Models with Negative Sampling: Unified Framework and Impact of Noise Distributions

预印本

On the Recoverability of Private Information Unlearning in Large Language Models

Beyond Penalization: Diffusion-based Out-of-Distribution Detection and Selective Regularization in Offline Reinforcement Learning

Learning Rules, Memorizing Facts: An Information-Theoretic View of Hallucination

The Capabilities and Limitations of Weak-to-Strong Generalization: Generalization and Calibration

On Weak-to-Strong Generalization and f-Divergence

Weak-to-Strong Generalization via Bregman Bias-Variance Decomposition

研究方向