arXiv subjects

Wei Zhuo

Publications and source records attributed to Wei Zhuo.

At least 19 recordsLinked to original sources

arXiv2026

CLAUSE: Agentic Neuro-Symbolic Knowledge Graph Reasoning via Dynamic Learnable Context Engineering

Knowledge graphs provide structured context for multi-hop question answering, but deployed systems must balance answer accuracy with strict latency and cost targets while preserving provenance. Static k-hop expansions and "think-longer" prompting often over-retrieve, inflate context, and yield unpredictable runtime. We introduce CLAUSE, an agentic three-agent neuro-symbolic framework that treats context construction as a sequential decision process over knowledge graphs, deciding what to expand, which paths to follow or backtrack, what evidence to keep, and when to stop. Latency (interaction steps) and prompt cost (selected tokens) are exposed as user-specified budgets or prices, allowing per-query adaptation to trade-offs among accuracy, latency, and cost without retraining. CLAUSE employs the proposed Lagrangian-Constrained Multi-Agent Proximal Policy Optimization (LC-MAPPO) algorithm to coordinate three agents: Subgraph Architect, Path Navigator, and Context Curator, so that subgraph construction, reasoning-path discovery, and evidence selection are jointly optimized under per-query resource budgets on edge edits, interaction steps, and selected tokens. Across HotpotQA, MetaQA, and FactKG, CLAUSE yields higher EM@1 while reducing subgraph growth and end-to-end latency at equal or lower token budgets. On MetaQA-2-hop, relative to the strongest RAG baseline (GraphRAG), CLAUSE achieves +39.3 EM@1 with 18.6% lower latency and 40.9% lower edge growth. The resulting contexts are compact, provenance-preserving, and deliver predictable performance under deployment constraints.

cs.AI↗

arXiv2026

An Efficient High-Degree, High-Order Equivariant Graph Neural Network for Direct Crystal Structure Optimization

Crystal structure optimization is fundamental to materials modeling but remains computationally expensive when performed with density-functional theory (DFT). Machine-learning (ML) approaches offer substantial acceleration, yet existing methods face three key limitations: (i) most models operate solely on atoms and treat lattice vectors implicitly, despite their central role in structural optimization; (ii) they lack efficient mechanisms to capture high-degree angular information and higher-order geometric correlations simultaneously, which are essential for distinguishing subtle structural differences; and (iii) many pipelines are multi-stage or iterative rather than truly end-to-end, making them prone to error accumulation and limiting scalability. Here we present E$^{3}$Relax-H$^{2}$, an end-to-end high-degree, high-order equivariant graph neural network that maps an initial crystal directly to its relaxed structure. The key idea is to promote both atoms and lattice vectors to graph nodes, enabling a unified and symmetry-consistent representation of structural degrees of freedom. Building on this formulation, E$^{3}$Relax-H$^{2}$ introduces two message-passing mechanisms: (i) a high-degree, high-order message-passing module that efficiently captures high-degree angular representations and high-order many-body correlations; and (ii) a lattice-atom message-passing module that explicitly models the bidirectional coupling between lattice deformation and atomic displacement. In addition, we propose a differentiable periodicity-aware Cartesian displacement loss tailored for one-shot structure prediction under periodic boundary conditions.

cond-mat.mtrl-sci↗

arXiv2026

Modality-free Graph In-context Alignment

In-context learning (ICL) converts static encoders into task-conditioned reasoners, enabling adaptation to new data from just a few examples without updating pretrained parameters. This capability is essential for graph foundation models (GFMs) to approach LLM-level generality. Yet current GFMs struggle with cross-domain alignment, typically relying on modality-specific encoders that fail when graphs are pre-vectorized or raw data is inaccessible. In this paper, we introduce Modality-Free Graph In-context Alignment (MF-GIA), a framework that makes a pretrained graph encoder promptable for few-shot prediction across heterogeneous domains without modality assumptions. MF-GIA captures domain characteristics through gradient fingerprints, which parameterize lightweight transformations that align pre-encoded features and indexed labels into unified semantic spaces. During pretraining, a dual prompt-aware attention mechanism with episodic objective learns to match queries against aligned support examples to establish prompt-based reasoning capabilities. At inference, MF-GIA performs parameter-update-free adaptation using only a few-shot support set to trigger cross-domain alignment and enable immediate prediction on unseen domains. Experiments demonstrate that MF-GIA achieves superior few-shot performance across diverse graph domains and strong generalization to unseen domains.

cs.LG↗

arXiv2025

UINO-FSS: Unifying Representation Learning and Few-shot Segmentation via Hierarchical Distillation and Mamba-HyperCorrelation

Few-shot semantic segmentation has attracted growing interest for its ability to generalize to novel object categories using only a few annotated samples. To address data scarcity, recent methods incorporate multiple foundation models to improve feature transferability and segmentation performance. However, they often rely on dual-branch architectures that combine pre-trained encoders to leverage complementary strengths, a design that limits flexibility and efficiency. This raises a fundamental question: can we build a unified model that integrates knowledge from different foundation architectures? Achieving this is, however, challenging due to the misalignment between class-agnostic segmentation capabilities and fine-grained discriminative representations. To this end, we present UINO-FSS, a novel framework built on the key observation that early-stage DINOv2 features exhibit distribution consistency with SAM's output embeddings. This consistency enables the integration of both models' knowledge into a single-encoder architecture via coarse-to-fine multimodal distillation. In particular, our segmenter consists of three core components: a bottleneck adapter for embedding alignment, a meta-visual prompt generator that leverages dense similarity volumes and semantic embeddings, and a mask decoder. Using hierarchical cross-model distillation, we effectively transfer SAM's knowledge into the segmenter, further enhanced by Mamba-based 4D correlation mining on support-query pairs. Extensive experiments on PASCAL-5$^i$ and COCO-20$^i$ show that UINO-FSS achieves new state-of-the-art results under the 1-shot setting, with mIoU of 80.6 (+3.8%) on PASCAL-5$^i$ and 64.5 (+4.1%) on COCO-20$^i$, demonstrating the effectiveness of our unified approach.

cs.CV↗

arXiv2025

Equivariant Atomic and Lattice Modeling Using Geometric Deep Learning for Crystal Structure Optimization

Structure optimization, which yields the relaxed structure (minimum-energy state), is essential for reliable materials property calculations, yet traditional ab initio approaches such as density-functional theory (DFT) are computationally intensive. Machine learning (ML) has emerged to alleviate this bottleneck but suffers from two major limitations: (i) existing models operate mainly on atoms, leaving lattice vectors implicit despite their critical role in structural optimization; and (ii) they often rely on multi-stage, non-end-to-end workflows that are prone to error accumulation. Here, we present E3Relax, an end-to-end equivariant graph neural network that maps an unrelaxed crystal directly to its relaxed structure. E3Relax promotes both atoms and lattice vectors to graph nodes endowed with dual scalar-vector features, enabling unified and symmetry-preserving modeling of atomic displacements and lattice deformations. A layer-wise supervision strategy forces every network depth to make a physically meaningful refinement, mimicking the incremental convergence of DFT while preserving a fully end-to-end pipeline. We evaluate E3Relax on four benchmark datasets and demonstrate that it achieves remarkable accuracy and efficiency. Through DFT validations, we show that the structures predicted by E3Relax are energetically favorable, making them suitable as high-quality initial configurations to accelerate DFT calculations.

cond-mat.mtrl-sci↗

arXiv2025

Introducing LongCat-Flash-Thinking: A Technical Report

We present LongCat-Flash-Thinking, an efficient 560-billion-parameter open-source Mixture-of-Experts (MoE) reasoning model. Its advanced capabilities are cultivated through a meticulously crafted training process, beginning with long Chain-of-Thought (CoT) data cold-start and culminating in large-scale Reinforcement Learning (RL). We first employ a well-designed cold-start training strategy, which significantly enhances the reasoning potential and equips the model with specialized skills in both formal and agentic reasoning. Then, a core innovation is our domain-parallel training scheme, which decouples optimization across distinct domains (e.g., STEM, Code, Agentic) and subsequently fuses the resulting expert models into a single, nearly Pareto-optimal model. This entire process is powered by our Dynamic ORchestration for Asynchronous rollout (DORA) system, a large-scale RL framework that delivers a greater than threefold training speedup over synchronous methods on tens of thousands of accelerators. As a result, LongCat-Flash-Thinking achieves state-of-the-art performance among open-source models on a suite of complex reasoning tasks. The model exhibits exceptional efficiency in agentic reasoning, reducing average token consumption by 64.5% (from 19, 653 to 6, 965) on AIME-25, without degrading task accuracy. We release LongCat-Flash-Thinking to promote further advances in reasoning systems and agentic AI research.

cs.AI↗

arXiv2025

IAD-GPT: Advancing Visual Knowledge in Multimodal Large Language Model for Industrial Anomaly Detection

The robust causal capability of Multimodal Large Language Models (MLLMs) hold the potential of detecting defective objects in Industrial Anomaly Detection (IAD). However, most traditional IAD methods lack the ability to provide multi-turn human-machine dialogues and detailed descriptions, such as the color of objects, the shape of an anomaly, or specific types of anomalies. At the same time, methods based on large pre-trained models have not fully stimulated the ability of large models in anomaly detection tasks. In this paper, we explore the combination of rich text semantics with both image-level and pixel-level information from images and propose IAD-GPT, a novel paradigm based on MLLMs for IAD. We employ Abnormal Prompt Generator (APG) to generate detailed anomaly prompts for specific objects. These specific prompts from the large language model (LLM) are used to activate the detection and segmentation functions of the pre-trained visual-language model (i.e., CLIP). To enhance the visual grounding ability of MLLMs, we propose Text-Guided Enhancer, wherein image features interact with normal and abnormal text prompts to dynamically select enhancement pathways, which enables language models to focus on specific aspects of visual data, enhancing their ability to accurately interpret and respond to anomalies within images. Moreover, we design a Multi-Mask Fusion module to incorporate mask as expert knowledge, which enhances the LLM's perception of pixel-level anomalies. Extensive experiments on MVTec-AD and VisA datasets demonstrate our state-of-the-art performance on self-supervised and few-shot anomaly detection and segmentation tasks, such as MVTec-AD and VisA datasets. The codes are available at \href{https://github.com/LiZeWen1225/IAD-GPT}{https://github.com/LiZeWen1225/IAD-GPT}.

cs.CV↗

arXiv2025

Personalized Subgraph Federated Learning with Differentiable Auxiliary Projections

Federated Learning (FL) on graph-structured data typically faces non-IID challenges, particularly in scenarios where each client holds a distinct subgraph sampled from a global graph. In this paper, we introduce Federated learning with Auxiliary projections (FedAux), a personalized subgraph FL framework that learns to align, compare, and aggregate heterogeneously distributed local models without sharing raw data or node embeddings. In FedAux, each client jointly trains (i) a local GNN and (ii) a learnable auxiliary projection vector (APV) that differentiably projects node embeddings onto a 1D space. A soft-sorting operation followed by a lightweight 1D convolution refines these embeddings in the ordered space, enabling the APV to effectively capture client-specific information. After local training, these APVs serve as compact signatures that the server uses to compute inter-client similarities and perform similarity-weighted parameter mixing, yielding personalized models while preserving cross-client knowledge transfer. Moreover, we provide rigorous theoretical analysis to establish the convergence and rationality of our design. Empirical evaluations across diverse graph benchmarks demonstrate that FedAux substantially outperforms existing baselines in both accuracy and personalization performance. The code is available at https://github.com/JhuoW/FedAux.

cs.LG↗

arXiv2025

Beyond Class Tokens: LLM-guided Dominant Property Mining for Few-shot Classification

Few-shot Learning (FSL), which endeavors to develop the generalization ability for recognizing novel classes using only a few images, faces significant challenges due to data scarcity. Recent CLIP-like methods based on contrastive language-image pertaining mitigate the issue by leveraging textual representation of the class name for unseen image discovery. Despite the achieved success, simply aligning visual representations to class name embeddings would compromise the visual diversity for novel class discrimination. To this end, we proposed a novel Few-Shot Learning (FSL) method (BCT-CLIP) that explores \textbf{dominating properties} via contrastive learning beyond simply using class tokens. Through leveraging LLM-based prior knowledge, our method pushes forward FSL with comprehensive structural image representations, including both global category representation and the patch-aware property embeddings. In particular, we presented a novel multi-property generator (MPG) with patch-aware cross-attentions to generate multiple visual property tokens, a Large-Language Model (LLM)-assistant retrieval procedure with clustering-based pruning to obtain dominating property descriptions, and a new contrastive learning strategy for property-token learning. The superior performances on the 11 widely used datasets demonstrate that our investigation of dominating properties advances discriminative class-specific representation learning and few-shot classification.

cs.CV↗

arXiv2025

3D Heart Reconstruction from Sparse Pose-agnostic 2D Echocardiographic Slices

Echocardiography (echo) plays an indispensable role in the clinical practice of heart diseases. However, ultrasound imaging typically provides only two-dimensional (2D) cross-sectional images from a few specific views, making it challenging to interpret and inaccurate for estimation of clinical parameters like the volume of left ventricle (LV). 3D ultrasound imaging provides an alternative for 3D quantification, but is still limited by the low spatial and temporal resolution and the highly demanding manual delineation. To address these challenges, we propose an innovative framework for reconstructing personalized 3D heart anatomy from 2D echo slices that are frequently used in clinical practice. Specifically, a novel 3D reconstruction pipeline is designed, which alternatively optimizes between the 3D pose estimation of these 2D slices and the 3D integration of these slices using an implicit neural network, progressively transforming a prior 3D heart shape into a personalized 3D heart model. We validate the method with two datasets. When six planes are used, the reconstructed 3D heart can lead to a significant improvement for LV volume estimation over the bi-plane method (error in percent: 1.98\% VS. 20.24\%). In addition, the whole reconstruction framework makes even an important breakthrough that can estimate RV volume from 2D echo slices (with an error of 5.75\% ). This study provides a new way for personalized 3D structure and function analysis from cardiac ultrasound and is of great potential in clinical practice.

eess.IV↗

arXiv2025

Commute Graph Neural Networks

Graph Neural Networks (GNNs) have shown remarkable success in learning from graph-structured data. However, their application to directed graphs (digraphs) presents unique challenges, primarily due to the inherent asymmetry in node relationships. Traditional GNNs are adept at capturing unidirectional relations but fall short in encoding the mutual path dependencies between nodes, such as asymmetrical shortest paths typically found in digraphs. Recognizing this gap, we introduce Commute Graph Neural Networks (CGNN), an approach that seamlessly integrates node-wise commute time into the message passing scheme. The cornerstone of CGNN is an efficient method for computing commute time using a newly formulated digraph Laplacian. Commute time is then integrated into the neighborhood aggregation process, with neighbor contributions weighted according to their respective commute time to the central node in each layer. It enables CGNN to directly capture the mutual, asymmetric relationships in digraphs. Extensive experiments on 8 benchmarking datasets confirm the superiority of CGNN against 13 state-of-the-art methods.

cs.LG↗

arXiv2025

EchoONE: Segmenting Multiple echocardiography Planes in One Model

In clinical practice of echocardiography examinations, multiple planes containing the heart structures of different view are usually required in screening, diagnosis and treatment of cardiac disease. AI models for echocardiography have to be tailored for each specific plane due to the dramatic structure differences, thus resulting in repetition development and extra complexity. Effective solution for such a multi-plane segmentation (MPS) problem is highly demanded for medical images, yet has not been well investigated. In this paper, we propose a novel solution, EchoONE, for this problem with a SAM-based segmentation architecture, a prior-composable mask learning (PC-Mask) module for semantic-aware dense prompt generation, and a learnable CNN-branch with a simple yet effective local feature fusion and adaption (LFFA) module for SAM adapting. We extensively evaluated our method on multiple internal and external echocardiography datasets, and achieved consistently state-of-the-art performance for multi-source datasets with different heart planes. This is the first time that the MPS problem is solved in one model for echocardiography data. The code will be available at https://github.com/a2502503/EchoONE.

cs.CV↗

arXiv2025

Optical Convolutional Spectrometer

Optical spectrometers are fundamental across numerous disciplines in science and technology. However, miniaturized versions, while essential for in situ measurements, are often restricted to coarse identification of signature peaks and inadequate for metrological purposes. Here, we introduce a new class of spectrometer, leveraging the convolution theorem as its mathematical foundation. Our convolutional spectrometer offers unmatched performance for miniaturized systems and distinct structural and computational simplicity, featuring a centimeter-scale footprint for the fully packaged unit, low cost (~$10) and a 2400 cm-1 (approximately 500 nm) bandwidth. We achieve excellent precision in resolving complex spectra with sub-second sampling and processing time, demonstrating a wide range of applications from industrial and agricultural analysis to healthcare monitoring. Specifically, our spectrometer system classifies diverse solid samples, including plastics, pharmaceuticals, coffee, flour and tea, with 100% success rate, and quantifies concentrations of aqueous and organic solutions with detection accuracy surpassing commercial benchtop spectrometers. We also realize the non-invasive sensing of human biomarkers, such as skin moisture (mean absolute error; MAE = 2.49%), blood alcohol (1.70 mg/dL), blood lactate (0.81 mmol/L), and blood glucose (0.36 mmol/L), highlighting the potential of this new class of spectrometers for low-cost, high-precision, portable/wearable spectral metrology.

physics.optics↗

arXiv2024

Partitioning Message Passing for Graph Fraud Detection

Label imbalance and homophily-heterophily mixture are the fundamental problems encountered when applying Graph Neural Networks (GNNs) to Graph Fraud Detection (GFD) tasks. Existing GNN-based GFD models are designed to augment graph structure to accommodate the inductive bias of GNNs towards homophily, by excluding heterophilic neighbors during message passing. In our work, we argue that the key to applying GNNs for GFD is not to exclude but to {\em distinguish} neighbors with different labels. Grounded in this perspective, we introduce Partitioning Message Passing (PMP), an intuitive yet effective message passing paradigm expressly crafted for GFD. Specifically, in the neighbor aggregation stage of PMP, neighbors with different classes are aggregated with distinct node-specific aggregation functions. By this means, the center node can adaptively adjust the information aggregated from its heterophilic and homophilic neighbors, thus avoiding the model gradient being dominated by benign nodes which occupy the majority of the population. We theoretically establish a connection between the spatial formulation of PMP and spectral analysis to characterize that PMP operates an adaptive node-specific spectral graph filter, which demonstrates the capability of PMP to handle heterophily-homophily mixed graphs. Extensive experimental results show that PMP can significantly boost the performance on GFD tasks.

cs.AI↗

arXiv2024

Graph Neural Networks with Feature and Structure Aware Random Walk

Graph Neural Networks (GNNs) have received increasing attention for representation learning in various machine learning tasks. However, most existing GNNs applying neighborhood aggregation usually perform poorly on the graph with heterophily where adjacent nodes belong to different classes. In this paper, we show that in typical heterphilous graphs, the edges may be directed, and whether to treat the edges as is or simply make them undirected greatly affects the performance of the GNN models. Furthermore, due to the limitation of heterophily, it is highly beneficial for the nodes to aggregate messages from similar nodes beyond local neighborhood.These motivate us to develop a model that adaptively learns the directionality of the graph, and exploits the underlying long-distance correlations between nodes. We first generalize the graph Laplacian to digraph based on the proposed Feature-Aware PageRank algorithm, which simultaneously considers the graph directionality and long-distance feature similarity between nodes. Then digraph Laplacian defines a graph propagation matrix that leads to a model called {\em DiglacianGCN}. Based on this, we further leverage the node proximity measured by commute times between nodes, in order to preserve the nodes' long-distance correlation on the topology level. Extensive experiments on ten datasets with different levels of homophily demonstrate the effectiveness of our method over existing solutions in the task of node classification.

cs.LG↗

arXiv2024

Chip-scale sensor for spectroscopic metrology

Miniaturized spectrometers hold great promise for in situ, in vitro, and even in vivo sensing applications. However, their size reduction imposes vital performance constraints in meeting the rigorous demands of spectroscopy, including fine resolution, high accuracy, and ultra-wide observation window. The prevailing view in the community holds that miniaturized spectrometers are most suitable for the coarse identification of signature peaks. In this paper, we present an integrated reconstructive spectrometer that enables near-infrared (NIR) spectroscopic metrology, and demonstrate a fully packaged sensor with auxiliary electronics. Such a sensor operates over a 520 nm bandwidth together with a resolution of less than 8 pm, which translates into a record-breaking bandwidth-to-resolution ratio of over 65,000. The classification of different types of solid substances and the concentration measurement of aqueous and organic solutions are performed, all achieving approximately 100% accuracy. Notably, the detection limit of our sensor matches that of the commercial benchtop counterparts, which is as low as 0.1% (i.e. 100 mg/dL) for identifying the concentration of glucose solution.

physics.optics↗

arXiv2024

Orthogonal Hyper-category Guided Multi-interest Elicitation for Micro-video Matching

Watching micro-videos is becoming a part of public daily life. Usually, user watching behaviors are thought to be rooted in their multiple different interests. In the paper, we propose a model named OPAL for micro-video matching, which elicits a user's multiple heterogeneous interests by disentangling multiple soft and hard interest embeddings from user interactions. Moreover, OPAL employs a two-stage training strategy, in which the pre-train is to generate soft interests from historical interactions under the guidance of orthogonal hyper-categories of micro-videos and the fine-tune is to reinforce the degree of disentanglement among the interests and learn the temporal evolution of each interest of each user. We conduct extensive experiments on two real-world datasets. The results show that OPAL not only returns diversified micro-videos but also outperforms six state-of-the-art models in terms of recall and hit rate.

cs.IR↗

arXiv2024

Efficient Graph Similarity Computation with Alignment Regularization

We consider the graph similarity computation (GSC) task based on graph edit distance (GED) estimation. State-of-the-art methods treat GSC as a learning-based prediction task using Graph Neural Networks (GNNs). To capture fine-grained interactions between pair-wise graphs, these methods mostly contain a node-level matching module in the end-to-end learning pipeline, which causes high computational costs in both the training and inference stages. We show that the expensive node-to-node matching module is not necessary for GSC, and high-quality learning can be attained with a simple yet powerful regularization technique, which we call the Alignment Regularization (AReg). In the training stage, the AReg term imposes a node-graph correspondence constraint on the GNN encoder. In the inference stage, the graph-level representations learned by the GNN encoder are directly used to compute the similarity score without using AReg again to speed up inference. We further propose a multi-scale GED discriminator to enhance the expressive ability of the learned representations. Extensive experiments on real-world datasets demonstrate the effectiveness, efficiency and transferability of our approach.

cs.LG↗