arXiv subjects

Ze Zhao

Publications and source records attributed to Ze Zhao.

At least 19 recordsLinked to original sources

arXiv2026

Inductive Reasoning for Temporal Knowledge Graphs with Emerging Entities

Reasoning on Temporal Knowledge Graphs (TKGs) is essential for predicting future events and time-aware facts. While existing methods are effective at capturing relational dynamics, their performance is limited by a closed-world assumption, which fails to account for emerging entities not present in the training. Notably, these entities continuously join the network without historical interactions. Empirical study reveals that emerging entities are widespread in TKGs, comprising roughly 25\% of all entities. The absence of historical interactions of these entities leads to significant performance degradation in reasoning tasks. Whereas, we observe that entities with semantic similarities often exhibit comparable interaction histories, suggesting the presence of transferable temporal patterns. Inspired by this insight, we propose TransFIR (Transferable Inductive Reasoning), a novel framework that leverages historical interaction sequences from semantically similar known entities to support inductive reasoning. Specifically, we propose a codebook-based classifier that categorizes emerging entities into latent semantic clusters, allowing them to adopt reasoning patterns from similar entities. Experimental results demonstrate that TransFIR outperforms all baselines in reasoning on emerging entities, achieving an average improvement of 28.6% in Mean Reciprocal Rank (MRR) across multiple datasets. The implementations are available at https://github.com/zhaodazhuang2333/TransFIR.

cs.AI↗

arXiv2026

Efficient Document Parsing via Parallel Token Prediction

Document parsing, as a fundamental yet crucial vision task, is being revolutionized by vision-language models (VLMs). However, the autoregressive (AR) decoding inherent to VLMs creates a significant bottleneck, severely limiting parsing speed. In this paper, we propose Parallel-Token Prediction (PTP), a plugable, model-agnostic and simple-yet-effective method that enables VLMs to generate multiple future tokens in parallel with improved sample efficiency. Specifically, we insert some learnable tokens into the input sequence and design corresponding training objectives to equip the model with parallel decoding capabilities for document parsing. Furthermore, to support effective training, we develop a comprehensive data generation pipeline that efficiently produces large-scale, high-quality document parsing training data for VLMs. Extensive experiments on OmniDocBench and olmOCR-bench demonstrate that our method not only significantly improves decoding speed (1.6x-2.2x) but also reduces model hallucinations and exhibits strong generalization abilities.

cs.CL↗

arXiv2025

CHAINSFORMER: Numerical Reasoning on Knowledge Graphs from a Chain Perspective

Reasoning over Knowledge Graphs (KGs) plays a pivotal role in knowledge graph completion or question answering systems, providing richer and more accurate triples and attributes. As numerical attributes become increasingly essential in characterizing entities and relations in KGs, the ability to reason over these attributes has gained significant importance. Existing graph-based methods such as Graph Neural Networks (GNNs) and Knowledge Graph Embeddings (KGEs), primarily focus on aggregating homogeneous local neighbors and implicitly embedding diverse triples. However, these approaches often fail to fully leverage the potential of logical paths within the graph, limiting their effectiveness in exploiting the reasoning process. To address these limitations, we propose ChainsFormer, a novel chain-based framework designed to support numerical reasoning. Chainsformer not only explicitly constructs logical chains but also expands the reasoning depth to multiple hops. Specially, we introduces Relation-Attribute Chains (RA-Chains), a specialized logic chain, to model sequential reasoning patterns. ChainsFormer captures the step-by-step nature of multi-hop reasoning along RA-Chains by employing sequential in-context learning. To mitigate the impact of noisy chains, we propose a hyperbolic affinity scoring mechanism that selects relevant logic chains in a variable-resolution space. Furthermore, ChainsFormer incorporates an attention-based numerical reasoner to identify critical reasoning paths, enhancing both reasoning accuracy and transparency. Experimental results demonstrate that ChainsFormer significantly outperforms state-of-the-art methods, achieving up to a 20.0% improvement in performance. The implementations are available at https://github.com/zhaodazhuang2333/ChainsFormer.

cs.AI↗

arXiv2025

KPIs 2024 Challenge: Advancing Glomerular Segmentation from Patch- to Slide-Level

Chronic kidney disease (CKD) is a major global health issue, affecting over 10% of the population and causing significant mortality. While kidney biopsy remains the gold standard for CKD diagnosis and treatment, the lack of comprehensive benchmarks for kidney pathology segmentation hinders progress in the field. To address this, we organized the Kidney Pathology Image Segmentation (KPIs) Challenge, introducing a dataset that incorporates preclinical rodent models of CKD with over 10,000 annotated glomeruli from 60+ Periodic Acid Schiff (PAS)-stained whole slide images. The challenge includes two tasks, patch-level segmentation and whole slide image segmentation and detection, evaluated using the Dice Similarity Coefficient (DSC) and F1-score. By encouraging innovative segmentation methods that adapt to diverse CKD models and tissue conditions, the KPIs Challenge aims to advance kidney pathology analysis, establish new benchmarks, and enable precise, large-scale quantification for disease research and diagnosis.

cs.CV↗

arXiv2024

Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent

In this paper, we introduce Hunyuan-Large, which is currently the largest open-source Transformer-based mixture of experts model, with a total of 389 billion parameters and 52 billion activation parameters, capable of handling up to 256K tokens. We conduct a thorough evaluation of Hunyuan-Large's superior performance across various benchmarks including language understanding and generation, logical reasoning, mathematical problem-solving, coding, long-context, and aggregated tasks, where it outperforms LLama3.1-70B and exhibits comparable performance when compared to the significantly larger LLama3.1-405B model. Key practice of Hunyuan-Large include large-scale synthetic data that is orders larger than in previous literature, a mixed expert routing strategy, a key-value cache compression technique, and an expert-specific learning rate strategy. Additionally, we also investigate the scaling laws and learning rate schedule of mixture of experts models, providing valuable insights and guidances for future model development and optimization. The code and checkpoints of Hunyuan-Large are released to facilitate future innovations and applications. Codes: https://github.com/Tencent/Hunyuan-Large Models: https://huggingface.co/tencent/Tencent-Hunyuan-Large

cs.CL↗

arXiv2024

Cross-Organ and Cross-Scanner Adenocarcinoma Segmentation using Rein to Fine-tune Vision Foundation Models

In recent years, significant progress has been made in tumor segmentation within the field of digital pathology. However, variations in organs, tissue preparation methods, and image acquisition processes can lead to domain discrepancies among digital pathology images. To address this problem, in this paper, we use Rein, a fine-tuning method, to parametrically and efficiently fine-tune various vision foundation models (VFMs) for MICCAI 2024 Cross-Organ and Cross-Scanner Adenocarcinoma Segmentation (COSAS2024). The core of Rein consists of a set of learnable tokens, which are directly linked to instances, improving functionality at the instance level in each layer. In the data environment of the COSAS2024 Challenge, extensive experiments demonstrate that Rein fine-tuned the VFMs to achieve satisfactory results. Specifically, we used Rein to fine-tune ConvNeXt and DINOv2. Our team used the former to achieve scores of 0.7719 and 0.7557 on the preliminary test phase and final test phase in task1, respectively, while the latter achieved scores of 0.8848 and 0.8192 on the preliminary test phase and final test phase in task2. Code is available at GitHub.

eess.IV↗

arXiv2024

OXYGENERATOR: Reconstructing Global Ocean Deoxygenation Over a Century with Deep Learning

Accurately reconstructing the global ocean deoxygenation over a century is crucial for assessing and protecting marine ecosystem. Existing expert-dominated numerical simulations fail to catch up with the dynamic variation caused by global warming and human activities. Besides, due to the high-cost data collection, the historical observations are severely sparse, leading to big challenge for precise reconstruction. In this work, we propose OxyGenerator, the first deep learning based model, to reconstruct the global ocean deoxygenation from 1920 to 2023. Specifically, to address the heterogeneity across large temporal and spatial scales, we propose zoning-varying graph message-passing to capture the complex oceanographic correlations between missing values and sparse observations. Additionally, to further calibrate the uncertainty, we incorporate inductive bias from dissolved oxygen (DO) variations and chemical effects. Compared with in-situ DO observations, OxyGenerator significantly outperforms CMIP6 numerical simulations, reducing MAPE by 38.77%, demonstrating a promising potential to understand the "breathless ocean" in data-driven manner.

cs.LG↗

arXiv2023

Graph Out-of-Distribution Generalization with Controllable Data Augmentation

Graph Neural Network (GNN) has demonstrated extraordinary performance in classifying graph properties. However, due to the selection bias of training and testing data (e.g., training on small graphs and testing on large graphs, or training on dense graphs and testing on sparse graphs), distribution deviation is widespread. More importantly, we often observe \emph{hybrid structure distribution shift} of both scale and density, despite of one-sided biased data partition. The spurious correlations over hybrid distribution deviation degrade the performance of previous GNN methods and show large instability among different datasets. To alleviate this problem, we propose \texttt{OOD-GMixup} to jointly manipulate the training distribution with \emph{controllable data augmentation} in metric space. Specifically, we first extract the graph rationales to eliminate the spurious correlations due to irrelevant information. Secondly, we generate virtual samples with perturbation on graph rationale representation domain to obtain potential OOD training samples. Finally, we propose OOD calibration to measure the distribution deviation of virtual samples by leveraging Extreme Value Theory, and further actively control the training distribution by emphasizing the impact of virtual OOD samples. Extensive studies on several real-world datasets on graph classification demonstrate the superiority of our proposed method over state-of-the-art baselines.

cs.LG↗

arXiv2023

Drag on a partially immersed sphere at the capillary scale

We study the drag on a centimetric sphere in a uniform flow in the presence of a free surface as a function of submergence depth. Through direct force measurements in a custom benchtop recirculating flume, we demonstrate that the drag can significantly exceed the corresponding drag in a single-phase flow and achieves a peak at submergence depths just prior to complete immersion. The additional drag in the partially immersed state is rationalized by considering hydrostatic effects associated with the asymmetric surface height profile induced by the obstacle in the flow direction which persists for flow speeds below the minimum capillary-gravity wave speed. At these scales, the sphere's wettability plays a pronounced role in determining the maximum possible drag and results in hysteretic behaviors near touchdown and complete immersion. The influence of flow speed, sphere size, and surface tension on the drag characteristics are additionally explored through a combination of experiments and numerical simulations.

physics.flu-dyn↗

arXiv2022

Computational thermal multi-phase flow for metal additive manufacturing

Thermal multi-phase flow simulations are indispensable to understanding the multi-scale and multi-physics phenomena in metal additive manufacturing (AM) processes, yet accurate and robust predictions remain challenging. This book chapter summarizes the recent method development at UIUC for simulating thermal multiphase flows in laser powder bed fusion (LPBF) and directed energy deposition (DED) processes. Two main method developments are discussed. The first is a mixed interface-capturing/interface-tracking computational framework aiming to explicitly treat the gas-metal interface without mesh motion/re-meshing. The second is a physics-based and non-empirical deposit geometry model for DED processes. The proposed framework's accuracy is assessed by thoroughly comparing the simulated results against experimental measurements on various quantities. We also report critical quantities that experiments can not measure to show the predictive capability of the developed methods.

cs.CE↗

arXiv2020

Theoretical interpretation of $\Xi_c(2970)$

The open charm strong decay widths and certain ratio of branching fractions of a charmed strange baryon $\Xi_c(2970)$ are calculated in a $^3P_0$ model. The results are compatible with the latest experimental data. The theoretical ratio of decay branching fractions $R = \mathcal{B} [ \Xi_{c}(2970)^+ \rightarrow \Xi_c(2645)^{0}\pi^{+} ] / \mathcal{B} [ \Xi_{c}(2970)^+ \rightarrow \Xi_c^{\prime0}\pi^{+}] \approx 1.0$. The spin-parity $J^P = 1/2^+$ and $3/2^+$ for different assignments are analyzed. From the results of our calculation, $\Xi_{c}(2970)$ can be interpreted as a $2S$-wave state with $J^P(s_l) = 1/2^+(0)$. The distinguishing between the 2S-wave $n_\rho$- and $n_\lambda$-excitation states and between states with $s_l = 0$ and $s_l = 1$ and between states with total spin $1/2$ and $3/2$($s_l = 1$) are also discussed.

hep-ph↗

arXiv2020

Dynamical mixing between $2^3S_1$ and $1^3D_1$ charmed mesons

In charmed $D$ and $D_s$ mesons sector, the matrix of a Hamiltonian in a quark potential model is computed in the $2^3S_1$ and $1^3D_1$ subspace. The masses of four mixed states of $2^3S_1$ and $1^3D_1$ denoted with $D^*_1(2635)$, $D^*_1(2739)$, $D^*_{s1}(2715)$ and $D^*_{s1}(2805)$ are obtained. It is an off-diagonal part of the spin-orbit tensor interaction that causes the mixing between the $2^3S_1$ and $1^3D_1$ states. The mixing angles between the $2^3S_1$ and $1^3D_1$ states are tiny. Under the mixing, a $^3P_0$ model is employed to compute the hadronic decay widths of all OZI-allowed decay channels of the four mixed states. The two light mixed states $D^*_1(2635)$ and $D^*_{s1}(2715)$ are close in mass to $D^*_J(2600)$ and $D^*_{s1}(2700)$, while the two heavy mixed states $D^*_1(2739)$ and $D^*_{s1}(2805)$ are lighter in mass than $D(2750)$ and $D^*_{s1}(2860)$. The mixing angles obtained from dynamical interaction are inconsistent with the mixing angles obtained from hadronic decay. Based on mass spectra and hadronic decay analyses, $D^*_J(2600)$, $D(2750)$, $D^*_{s1}(2700)$ and $D^*_{s1}(2860)$ are impossibly the mixed states of $2^3S_1$ and $1^3D_1$ at the small mixing angles. The inconsistence implies that $D^*_1(2760)$ and $D^*_{s1}(2860)$ have not been properly resolved from present experimental data, or there exist large unknown off-diagonal interactions that result in large mixing angles.

hep-ph↗

arXiv2017

Study of $2S$- and $1D$- excitations of observed charmed strange baryons

Strong decays of $\Xi_c$ baryons with radial or orbital excitations with positive parity have been studied in a $^3P_0$ model. As candidates of these $Xi_c$, possible configurations and $J^P$ of $\Xi_c(2930)$, $\Xi_c(2980)$, $\Xi_c(3055)$, $\Xi_c(3080)$ and $\Xi_c(3123)$ have been assigned. There are $40$ kinds of configurations to describe these excited $\Xi_c$. In these assignments, $\Xi_c(2930)$ may be a $2S$-wave excited $\tilde{\Xi}_{c1}(\frac{1}{2}^+)$ or $\tilde{\Xi}_{c1}(\frac{3}{2}^+)$, or a $D$-wave excited $\hat\Xi_{c1}^{' }(\frac{1}{2}^+)$, $\check\Xi_{c1}^{\ 0}(\frac{1}{2}^+)$, $\check\Xi_{c1}^{\ 2}(\frac{1}{2}^+)$, $\hat\Xi_{c1}^{' }(\frac{3}{2}^+)$, $\check\Xi_{c1}^{\ 0}(\frac{3}{2}^+)$ or $\check\Xi_{c1}^{\ 2}(\frac{3}{2}^+)$. $\Xi_c(2980)^+$ may be a $2S$-wave excited $\tilde{\Xi}_{c1}(\frac{1}{2}^+)$ or $\tilde{\Xi}_{c0}^{'}(\frac{1}{2}^+)$ with $J^P={1\over 2}^+$, or a $D$-wave excited $\check\Xi_{c0}^{'0}(\frac{1}{2}^+)$ or $\check\Xi_{c1}^{\ 0}(\frac{1}{2}^+)$ with $J^P={1\over 2}^+$. $\Xi_c(3055)^+$ may be a $2S$-wave excited $\acute{\Xi}_{c1}^{'}(\frac{3}{2}^+)$ or $\acute{\Xi}_{c0}(\frac{1}{2}^+)$. It may be a $D$-wave excited $\Xi_{c1}^{' }(\frac{3}{2}^+)$, $\Xi_{c2}^{' }(\frac{5}{2}^+)$, $\Xi_{c2}^{ }(\frac{3}{2}^+)$ or $\Xi_{c2}^{ }(\frac{5}{2}^+)$. $\Xi_c(3080)^+$ is very possibly a $2S$-wave excited $\acute{\Xi}_{c0}(\frac{1}{2}^+)$, and seems not a $D$-wave excitation of $\Xi_c$. For the poor experimental information of $\Xi_c(3123)$, it is impossible to identify this state at present. It is found that the channel $\Lambda D$ vanishes in the strong decay of $P$-wave, $D$-wave and $2S$-wave excited $\Xi_c$ without $\rho$- mode excitation between the two light quarks ($n_\rho=L_\rho=0$). Some branching fraction ratios have been computed and can be employed to distinguish different configurations in forthcoming experiments.

hep-ph↗

arXiv2017

Diagnosing holographic type dark energy models with the Statefinder hierarchy, composite null diagnostic and $w-w'$ pair

The main purpose of this work is to distinguish various holographic type dark energy (DE) models, including the $\Lambda$HDE, HDE, NADE and RDE model, by using various diagnostic tools. The first diagnostic tool is the Statefinder hierarchy, in which the evolution of Statefinder hierarchy parmeter $S^{(1)}_3(z)$ and $S^{(1)}_4(z)$ are studied. The second is composite null diagnostic (CND), in which the trajectories of $\{S^{(1)}_3, \epsilon\}$ and $\{S^{(1)}_4, \epsilon\}$ are investigated, where $\epsilon$ is the fractional growth parameter. The last is $w-w'$ analysis, where $w$ is the equation of state for DE and the prime denotes derivative with respect to $ln a$. In the analysis we consider two cases: varying current fractional DE density $\Omega_{de0}$ and varying DE model parameter $C$. We find that: (1) Both the Statefinder hierarchy and the CND have qualitative impact on $\Lambda$HDE, but only have quantitative impact on HDE. (2) $S_4^{(1)}$ can lead to larger differences than $S_3^{(1)}$, while the CND pair has a stronger ability to distinguish different models than the Statefinder hierarchy. (3) For the case of varying $C$, the $\{w, w'\}$ pair has qualitative impact on $\Lambda$HDE; for the case of varying $\Omega_{de0}$, the $\{w, w'\}$ pair only has quantitative impact; these results are different from the cases of HDE, RDE and NADE, in which the $\{w, w'\}$ pair only has quantitative impact on these models. In conclusion, compared with HDE, RDE and NADE, the $\Lambda$HDE model can be easily distinguished by using these diagnostic tools.

astro-ph.CO↗

arXiv2017

Study of $P$-wave excitations of observed charmed strange baryons

Many excited charmed strange baryons such as $\Xi_c(2790)$, $\Xi_c(2815)$, $\Xi_c(2930)$, $\Xi_c(2980)$, $\Xi_c(3055)$, $\Xi_c(3080)$ and $\Xi_c(3123)$ have been observed. In order to understand their internal structure and to determine their spin-parities, the strong decay properties of these baryons as possible $P$-wave excited $\Xi_c$ candidates have been systematically studied in a $^3P_0$ model. The configurations and $J^P$ assignments of $\Xi_c(2790)$, $\Xi_c(2815)$, $\Xi_c(2930)$, $\Xi_c(2980)$, $\Xi_c(3055)$, $\Xi_c(3080)$ and $\Xi_c(3123)$ have been explored based on recent experimental data. In our analyses, $\Xi_c(3055)$, $\Xi_c(3080)$ and $\Xi_c(3123)$ seem impossible to be the $P$-wave excited $\Xi_c$. $\Xi_c(2790)$, $\Xi_c(2815)$, $\Xi_c(2930)$ and $\Xi_c(2980)$ may be the $P$-wave excited $\Xi_c$. In particular, $\Xi_c(2790)$ and $\Xi_c(2815)$ are very possibly the $P$-wave excited $\Xi_{c1}(1/2^-)$ and $\Xi_{c1}(3/2^-)$, respectively. $\Xi_c(2980)$ may be the $P$-wave excited $\Xi_{c1}^{'}(\frac{1}{2}^-)$. $\Xi_c(2930)$ may be the $P$-wave $\Xi_{c0}^{'}(\frac{1}{2}^-)$, $\tilde{\Xi}_{c0}(\frac{1}{2}^-)$, $\Xi_{c2}^{'}(\frac{3}{2}^-)$, $\Xi_{c2}^{'}(\frac{5}{2}^-)$, $\tilde{\Xi}_{c2}(\frac{3}{2}^-)$ or $\tilde{\Xi}_{c2}(\frac{5}{2}^-)$. Furthermore, some branching fraction ratios related to the internal structure and quark configuration of $P$-wave $\Xi_c$ have also been computed. Measurements of these ratios in the future will be helpful to understand these excited $\Xi_c$.

hep-ph↗

arXiv2017

Hadronic decay properties of newly observed $\Omega_c$ baryons

Hadronic decay widths of the newly observed charmed strange baryons, $\Omega_c(3000)^0$, $\Omega_c(3050)^0$, $\Omega_c(3066)^0$, $\Omega_c(3090)^0$ and $\Omega_c(3119)^0$ have been calculated in a $^3P_0$ model. Our results indicate that $\Omega_c(3066)^0$ and $\Omega_c(3090)^0$ can be interpreted as the $1P-$wave $\Omega_{c2}(\frac{3}{2}^-)$ or $\Omega_{c2}(\frac{5}{2}^-)$. Though the measured masses of $\Omega_c(3000)^0$, $\Omega_c(3050)^0$ and $\Omega_c(3119)^0$ are lower than existed theoretical predictions of $1D-$wave $\Omega_c$, the hadronic decay features of these $\Omega_c$ favor assignments of the $1D-$wave states. $\Omega_c(3000)^0$ is possibly $\Omega_{c1}(\frac{1}{2}^+)$ or $\Omega_{c1}(\frac{3}{2}^+)$, $\Omega_c(3050)^0$ is possibly $\hat\Omega_{c3}(\frac{5}{2}^+)$ or $\hat\Omega_{c3}(\frac{7}{2}^+)$, and $\Omega_c(3119)^0$ is possibly $\hat\Omega_{c3}(\frac{5}{2}^+)$, $\hat\Omega_{c3}(\frac{7}{2}^+)$, $\Omega_{c3}(\frac{5}{2}^+)$ or $\Omega_{c3}(\frac{7}{2}^+)$. The predicted total decay widths in these assignments are consistent with experiment.

hep-ph↗

arXiv2017

Study of radially excited $D_s(2^1S_0)$ and $D^+_{sJ}(2632)$

The $J^P=0^-$ radial excitation $D_s(2^1S_0)$ is anticipated to have mass $2650$ MeV (denoted with $D_s(2650)$) though it has not been observed. $D_s(2650)$ is anticipated to be observed in inclusive $e^+e^-$ and $pp$ collisions in $D^*K$ channel. $D_s(2650)$ can be produced from hadronic decays of higher excited resonances. In a $^3P_0$ model, hadronic production of $D_s(2650)$ has been studied, relevant decay widths have been estimated. The hadronic decays of $D_s(3P)$ have been explored in detail, and the dominant decay channels have been pointed out. Two hadronic decay channels of $D^+_{sJ}(2632)$ were observed by SELEX, but have not been observed by any other experiment. Hadronic decays of $D^+_{sJ}(2632)$ in different assignments have been explored in the $^3P_0$ model. In view of the hadronic decay manner of $D^+_{sJ}(2632)$ , we conclude that $D^+_{sJ}(2632)$ cannot be a conventional charmed strange $c\bar s$ meson.

hep-ph↗

arXiv2016

Hadronic production of $D(2550)$, $D^(2600)$, $D(2750)$, $D^_1(2760)$ and $D^*_3(2760)$

Hadronic decays of radially excited $2D$ $D(2^3D_1)$, $D(2^3D_3)$, $D(2D_2)$ and $D(2D^\prime_2)$ have been studied in a $^3P_0$ model. All OZI-allowed decay channels of these $2D$ $D$ resonances have been given, and relevant decay widths have been calculated. $D(2550)$, $D^*(2600)$, $D(2750)$, $D^*_1(2760)$ and $D^*_3(2760)$ can be produced in hadronic decays of $D(2^3D_1)$, $D(2^3D_3)$, $D(2D_2)$ and $D(2D^\prime_2)$. In different assignments, hadronic decay widths and some relevant ratios from the $2D$ $D$ resonances to $D(2550)$, $D^*(2600)$, $D(2750)$, $D^*_1(2760)$ or $D^*_3(2760)$ final states have been predicted, which may provide some more information to identify these resonances in forthcoming experiments.

hep-ph↗