arXiv subjects

Yiqing Shi

Publications and source records attributed to Yiqing Shi.

6 recordsLinked to original sources

arXiv2026

Knockoffs-based False Discovery Rate Control and Simplification for Deep Neural Networks

The deep neural network is a widely used framework in machine learning that has been widely applied in various fields. However, deep neural networks often involve a large number of parameters and inputs, many of which may be irrelevant to the goal or true output. These parameters and input variables not only increase computational complexity, but also contribute to additional computational cost. One solution to this problem is knockoff methods, which have proven successful in controlling false discovery rates in high-dimensional regression. Building on the knockoff methods and using the regularised neural network, this paper proposes three variable screening methods under the condition of controlling false discovery rates: one layer filter, multiple layers filter, and variable weight aggregation filter. In comparison with existing algorithms, we find that our algorithms show satisfactory performance.

stat.ML↗

arXiv2026

SoK: Systematizing Software Artifacts Traceability via Associations, Techniques, and Applications

Software development relies heavily on traceability links between various software artifacts to ensure quality and facilitate maintenance. While automated traceability recovery techniques have advanced for different artifact pairs, the field remains fragmented with an incomplete overview of artifact associations, ambiguous linking techniques, and fragmented knowledge of application scenarios. To bridge these gaps, we conducted a systematic literature review on software traceability recovery to synthesize the linked artifacts, recovery tools, and usage scenarios across the traceability ecosystem. First, we constructed the first global artifacts traceability graph of 23 associations among 22 artifact types, exposing a severe research imbalance that heavily favors code-related links. Second, while recovery techniques are shifting toward deep semantic models, a reproducibility crisis persists (e.g., only 37% of studies released code); to address this, we provided a comprehensive evaluation framework including a technical decision map and standardized benchmarks. Finally, we quantified an industrial adoption gap (i.e., 95% of tools remain confined to academia) and proposed a role-centric framework to dynamically align artifact paths with concrete engineering activities. This review contributes a coherent knowledge framework for artifacts traceability research, identifies current trends, and provides directions for future work.

cs.SE↗

arXiv2025

Edit2Perceive: Image Editing Diffusion Models Are Strong Dense Perceivers

Recent advances in diffusion transformers have shown remarkable generalization in visual synthesis, yet most dense perception methods still rely on text-to-image (T2I) generators designed for stochastic generation. We revisit this paradigm and show that image editing diffusion models are inherently image-to-image consistent, providing a more suitable foundation for dense perception task. We introduce Edit2Perceive, a unified diffusion framework that adapts editing models for depth, normal, and matting. Built upon the FLUX.1 Kontext architecture, our approach employs full-parameter fine-tuning and a pixel-space consistency loss to enforce structure-preserving refinement across intermediate denoising states. Moreover, our single-step deterministic inference yields up to faster runtime while training on relatively small datasets. Extensive experiments demonstrate comprehensive state-of-the-art results across all three tasks, revealing the strong potential of editing-oriented diffusion transformers for geometry-aware perception.

cs.CV↗

arXiv2025

The emergence of nonlinear Jeans-type instabilities for quasilinear wave equations. II: Generalizations

This work extends the previous work by the first author [arXiv:2409.02516] and [Math. Ann. 393 (2025), 317-363], analyzing the long-term behavior of solutions to a broader class of quasilinear wave equations with parameter $1<\mathsf{a}\leq30$ and $\frac{1}{3}\leq\mathsf{b}\leq\frac{2}{3}$: \begin{equation*} \partial^2_t \varrho- \biggl( \frac{ \mathsf{m}^2 (\partial_{t}\varrho )^2}{(1+\varrho )^2} + 4(\mathsf{k}-\mathsf{m}^2)(1+\varrho )\biggr) \Delta \varrho = F(t,\varrho,\partial_{\mu} \varrho) \end{equation*} where $F$ is given by \begin{equation*} F(t,\varrho,\partial_{\mu} \varrho):= \mathsf{b} \varrho (1+ \varrho ) -(\mathsf{a}-1) \partial_{t}\varrho + \frac{4}{3} \frac{(\partial_{t}\varrho )^2}{1+\varrho } + \biggl(\mathsf{m}^2 \frac{ (\partial_{t}\varrho )^2}{(1+\varrho )^2} + 4(\mathsf{k}-\mathsf{m}^2) (1+\varrho ) \biggr) q^i \partial_{i}\varrho - \mathtt{K}^{ij} \partial_{i}\varrho\partial_{j}\varrho . \end{equation*} The results demonstrate that for this extensive family of quasilinear wave equations satisfying $1<\mathsf{a}\leq30$ and $\frac{1}{3}\leq\mathsf{b}\leq\frac{2}{3}$, self-increasing blowup solutions also exist, and self-increasing singularities emerge at certain future endpoints of null geodesics provided the inhomogeneous perturbations of data are sufficiently small.

math.AP↗

arXiv2023

Yelp Reviews and Food Types: A Comparative Analysis of Ratings, Sentiments, and Topics

This study examines the relationship between Yelp reviews and food types, investigating how ratings, sentiments, and topics vary across different types of food. Specifically, we analyze how ratings and sentiments of reviews vary across food types, cluster food types based on ratings and sentiments, infer review topics using machine learning models, and compare topic distributions among different food types. Our analyses reveal that some food types have similar ratings, sentiments, and topics distributions, while others have distinct patterns. We identify four clusters of food types based on ratings and sentiments and find that reviewers tend to focus on different topics when reviewing certain food types. These findings have important implications for understanding user behavior and cultural influence on digital media platforms and promoting cross-cultural understanding and appreciation.

cs.CL↗

arXiv2022

Rigorous proof of slightly nonlinear Jeans instability in the expanding Newtonian universe

Due to the nonlinearity of the Euler{Poisson equations, it is possible that the nonlinear Jeans instability may lead to a faster density growing rate than the rate in the standard theory of linearized Jeans instability, which motivates us to study the nonlinear Jeans instability. The aim of this article is to develop a method proving the Jeans instability for slightly nonlinear Euler-Poisson equations in the expanding Newtonian universe. The standard proofs of the Jeans instability rely on the Fourier analysis. However, it is difficult to generalize Fourier method to a nonlinear setting, and thus there is no result in the nonlinear analysis of Jeans instability. We firstly develop a non-Fourier-based method to reprove the linearized Jeans instability in the expanding Newtonian universe. Secondly, we generalize this idea to a slightly nonlinear case. This method relies on the Cauchy problem of the Fuchsian system due to the recent developments of this system in mathematics. The fully nonlinear Jeans instability for the Euler-Poisson and Einstein-Euler equations are in progress.

math.AP↗