BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Yibo Gao

Yibo Gao appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
math.COComputer Visioneess.IVmath.GRmath.RT

Catalog footprint

What is connected

13works
5topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

13 published item(s)

preprint2026arXiv

Beyond Forgetting in Continual Medical Image Segmentation: A Comprehensive Benchmark Study

Continual learning (CL) is essential for deploying medical image segmentation models in clinical environments where imaging domains, anatomical targets, and diagnostic tasks evolve over time. However, continual segmentation still faces three main challenges. First, the scenarios for this task remain insufficiently standardized for real-world clinical settings. Second, existing research has been primarily focused on mitigating forgetting, overlooking the other essential properties such as plasticity. Third, a benchmark work with comprehensive evaluation on existing methods is stll desirable. To address these gaps, we present such benchmark study of continual medical image segmentation. We first define three clinically motivated scenarios, namely Domain-CL, Class-CL, and Organ-CL, to respectively capture the cross-center domain shift, the incremental anatomical structure segmentation, and the cross-organ segmentation. We then introduce an evaluation framework that measures not only general performance and forgetting, but also plasticity, forward generalizability, parameter efficiency, and replay burden. The results, from extensive experiments with representative CL methods, showed that it was still challenging to develop a model that could satisfy all the requirements simultaneously. Nevertheless, these studies also suggested that the replay-based methods achieve the best overall balance between stability and plasticity, the parameter-isolation methods should be effective at reducing forgetting, though at the cost of increased model size, and the forward generalizability remain a significantly understudied aspect of this research field. Finally, we discuss related learning paradigms and outline future directions for continual medical image segmentation.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Joint Modeling of Image and Label Statistics for Enhancing Model Generalizability of Medical Image Segmentation

Although supervised deep-learning has achieved promising performance in medical image segmentation, many methods cannot generalize well on unseen data, limiting their real-world applicability. To address this problem, we propose a deep learning-based Bayesian framework, which jointly models image and label statistics, utilizing the domain-irrelevant contour of a medical image for segmentation. Specifically, we first decompose an image into components of contour and basis. Then, we model the expected label as a variable only related to the contour. Finally, we develop a variational Bayesian framework to infer the posterior distributions of these variables, including the contour, the basis, and the label. The framework is implemented with neural networks, thus is referred to as deep Bayesian segmentation. Results on the task of cross-sequence cardiac MRI segmentation show that our method set a new state of the art for model generalizability. Particularly, the BayeSeg model trained with LGE MRI generalized well on T2 images and outperformed other models with great margins, i.e., over 0.47 in terms of average Dice. Our code is available at https://zmiclab.github.io/projects.html.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Towards Self-supervised and Weight-preserving Neural Architecture Search

Neural architecture search (NAS) algorithms save tremendous labor from human experts. Recent advancements further reduce the computational overhead to an affordable level. However, it is still cumbersome to deploy the NAS techniques in real-world applications due to the fussy procedures and the supervised learning paradigm. In this work, we propose the self-supervised and weight-preserving neural architecture search (SSWP-NAS) as an extension of the current NAS framework by allowing the self-supervision and retaining the concomitant weights discovered during the search stage. As such, we simplify the workflow of NAS to a one-stage and proxy-free procedure. Experiments show that the architectures searched by the proposed framework achieve state-of-the-art accuracy on CIFAR-10, CIFAR-100, and ImageNet datasets without using manual labels. Moreover, we show that employing the concomitant weights as initialization consistently outperforms the random initialization and the two-stage weight pre-training method by a clear margin under semi-supervised learning scenarios. Codes are publicly available at https://github.com/LzVv123456/SSWP-NAS.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Counting Shellings of Complete Bipartite Graphs and Trees

A shelling of a graph, viewed as an abstract simplicial complex that is pure of dimension 1, is an ordering of its edges such that every edge is adjacent to some other edges appeared previously. In this paper, we focus on complete bipartite graphs and trees. For complete bipartite graphs, we obtain an exact formula for their shelling numbers. And for trees, we relate their shelling numbers to linear extensions of tree posets and bound shelling numbers using vertex degrees and diameter.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

A combinatorial duality between the weak and strong Bruhat orders

In recent work, the authors used an order lowering operator $\nabla$, introduced by Stanley, to prove the strong Sperner property for the weak Bruhat order on the symmetric group. Hamaker, Pechenik, Speyer, and Weigandt interpreted $\nabla$ as a differential operator on Schubert polynomials and used this to prove a new identity for Schubert polynomials and a determinant conjecture of Stanley. In this paper we study a raising operator $Δ$ for the \emph{strong} Bruhat order, which is in many ways dual to $\nabla$. We prove a Schubert identity dual to that of Hamaker et al. and derive formulas for counting weighted paths in the Hasse diagrams of the strong order which agree with path counting formulas for the weak order. We also show that powers of $\nabla$ and $Δ$ have the same Smith normal forms, which we describe explicitly, answering a question of Stanley.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Boolean elements in the Bruhat order

We show that $w\in W$ is boolean if and only if it avoids a set of Billey-Postnikov patterns, which we describe explicitly. Our proof is based on an analysis of inversion sets, and it is in large part type-uniform. We also introduce the notion of linear pattern avoidance, and show that boolean elements are characterized by avoiding just the $3$ linear patterns $s_1 s_2 s_1 \in W(A_2)$, $s_2 s_1 s_3 s_2 \in W(A_3)$, and $s_2 s_1 s_3 s_4 s_2 \in W(D_4)$. We also consider the more general case of $k$-boolean Weyl group elements. We say that $w\in W$ is $k$-boolean if every reduced expression for $w$ contains at most $k$ copies of each generator. We show that the $2$-boolean elements of the symmetric group $S_n$ are characterized by avoiding the patterns $3421,4312,4321,$ and $456123$, and give a rational generating function for the number of $2$-boolean elements of $S_n$.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Compatible Recurrent Identities of the Sandpile Group and Maximal Stable Configurations

In the abelian sandpile model, recurrent chip configurations are of interest as they are a natural choice of coset representatives under the quotient of the reduced Laplacian. We investigate graphs whose recurrent identities with respect to different sinks are compatible with each other. The maximal stable configuration is the simplest recurrent chip configuration, and graphs whose recurrent identities equal the maximal stable configuration are of particular interest, and are said to have the complete maximal identity property. We prove that given any graph $G$ one can attach trees to the vertices of $G$ to yield a graph with the complete maximal identity property. We conclude with several intriguing conjectures about the complete maximal identity property of various graph products.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Separable elements in Weyl groups

We define the notion of a separable element in a finite Weyl group, generalizing the well-studied class of separable permutations. We prove that the upper and lower order ideals in weak Bruhat order generated by a separable element are rank-symmetric and rank-unimodal, and that the product of their rank generating functions gives that of the whole group, answering an open problem of Fan Wei. We also prove that separable elements are characterized by pattern avoidance in the sense of Billey and Postnikov.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

The Largest Pure Partial Planes of Order 6 Have Size 25

In this paper, we prove that the largest pure partial plane of order 6 has size 25. At the same time, we classify all pure partial planes of order 6 and size 25 up to isomorphism. Our major approach is computer search. The search space is very large so we use combinatorial arguments to rule out some of the cases. For the remaining cases, we subdivide each search by phases and use multiple checks to reduce search space via symmetry.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Toric Mutations in the dP$_2$ Quiver and Subgraphs of the dP$_2$ Brane Tiling

Brane tilings are infinite, bipartite, periodic, planar graphs that are dual to quivers. In this paper, we examine the del Pezzo 2 (dP$_2$) quiver and its brane tiling, which arise from the physics literature, in terms of toric mutations on its corresponding cluster. Specifically, we give explicit formulas for all cluster variables generated by toric mutation sequences. Moreover, for each such variable, we associate a subgraph of the dP$_2$ brane tiling to it such that its weight matches the variable.

0 citations0 reviewsNo code linkedOpen access