1 citations · 1 across the 10 of their papers we have counts for
13 papers
The Berry-Esseen Constant Conjecture is Eventually True
Hengzhi He, Guang Cheng
We prove that the Esseen's constant is valid in the iid Berry--Esseen inequality as long as the sample sizes . The threshold is universal over all real summ…
Fractional Triangle Decompositions at Partite Minimum Degree
Hengzhi He, Guang Cheng
We prove that every triangle-divisible balanced tripartite graph whose vertex classes have size and whose partite minimum degree is at least admits a fractional triangle…
Tensor-normal maximum likelihood estimation at the operator-norm sample threshold
Hengzhi He, Guang Cheng
Let be independent Gaussian tensors in with a common covariance matrix given by the Kronecker product of …
Asymptotically Lossless Exact and Pivotal E-Values
Hengzhi He, Guang Cheng
Consider testing a finite composite null $\cP={P_1,\ldots,P_L}$ against a simple alternative using i.i.d. observations. Let denote the supremum of the expected log…
Sharp proper estimation of fixed-component Gaussian location mixtures in polynomial time
Hengzhi He, Guang Cheng
We consider a mixture of at most unit-covariance Gaussians in whose means belong to a fixed-radius ball, with no separation or minimum-weight condition. Doss, Wu…
Let the Target Select for Itself: Data Selection via Target-Aligned Paths
Huitao Yang, Hengzhi He, Guang Cheng +1
Targeted data selection seeks training examples from a candidate pool that improve downstream task performance. While trajectory-based selectors effectively guide this process, exi…
Authenticated Contradictions from Desynchronized Provenance and Watermarking
Alexander Nemecek, Hengzhi He, Guang Cheng +1
Cryptographic provenance standards such as C2PA and invisible watermarking are positioned as complementary defenses for content authentication, yet the two verification layers are…
Optimal Watermark Generation under Type I and Type II Errors
Hengzhi He, Shirong Xu, Alexander Nemecek +3
Watermarking has recently emerged as a crucial tool for protecting the intellectual property of generative models and for distinguishing AI-generated content from human-generated d…
A Probabilistic Perspective on Model Collapse
Shirong Xu, Hengzhi He, Guang Cheng
In recent years, model collapse has become a critical issue in language model training, making it essential to understand the underlying mechanisms driving this phenomenon. In this…
Recursive Learning Without Collapse: A Weighting-Based Stabilization Framework
Hengzhi He, Shirong Xu, Guang Cheng
Recent studies identified an intriguing phenomenon in recursive generative model training known as model collapse, where models trained on data generated by previous models exhibit…
Watermarking Generative Categorical Data
Bochao Gu, Hengzhi He, Guang Cheng
In this paper, we propose a novel statistical framework for watermarking generative categorical data. Our method systematically embeds pre-agreed secret signals by splitting the da…
Watermarking Continuous Tabular Data Generation
Hengzhi He, Peiyu Yu, Junpeng Ren +2
In this paper, we introduce a simple yet effective tabular data watermarking mechanism with statistical guarantees. We show theoretically that the proposed watermark can be effecti…
"Noisier" Noise Contrastive Eestimation is (Almost) Maximum Likelihood
Peiyu Yu, Dinghuai Zhang, Hengzhi He +10
Noise Contrastive Estimation (NCE) has fueled major breakthroughs in representation learning and generative modeling. Yet a long-standing challenge remains: accurately estimating r…