arXiv subjects

Yue Wu

Publications and source records attributed to Yue Wu.

At least 109 records · Page 6Linked to original sources

arXiv2025

DeXposure: A Dataset and Benchmarks for Inter-protocol Credit Exposure in Decentralized Financial Networks

We curate the DeXposure dataset, the first large-scale dataset for inter-protocol credit exposure in decentralized financial networks, covering global markets of 43.7 million entries across 4.3 thousand protocols, 602 blockchains, and 24.3 thousand tokens, from 2020 to 2025. A new measure, value-linked credit exposure between protocols, is defined as the inferred financial dependency relationships derived from changes in Total Value Locked (TVL). We develop a token-to-protocol model using DefiLlama metadata to infer inter-protocol credit exposure from the token's stock dynamics, as reported by the protocols. Based on the curated dataset, we develop three benchmarks for machine learning research with financial applications: (1) graph clustering for global network measurement, tracking the structural evolution of credit exposure networks, (2) vector autoregression for sector-level credit exposure dynamics during major shocks (Terra and FTX), and (3) temporal graph neural networks for dynamic link prediction on temporal graphs. From the analysis, we observe (1) a rapid growth of network volume, (2) a trend of concentration to key protocols, (3) a decline of network density (the ratio of actual connections to possible connections), and (4) distinct shock propagation across sectors, such as lending platforms, trading exchanges, and asset management protocols. The DeXposure dataset and code have been released publicly. We envision they will help with research and practice in machine learning as well as financial risk monitoring, policy analysis, DeFi market modeling, amongst others. The dataset also contributes to machine learning research by offering benchmarks for graph clustering, vector autoregression, and temporal graph analysis.

cs.LG↗

arXiv2025

FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning

Federated learning (FL) enables collaborative model training across multiple parties without sharing raw data, with semi-asynchronous FL (SAFL) emerging as a balanced approach between synchronous and asynchronous FL. However, SAFL faces significant challenges in optimizing both gradient-based (e.g., FedSGD) and model-based (e.g., FedAvg) aggregation strategies, which exhibit distinct trade-offs in accuracy, convergence speed, and stability. While gradient aggregation achieves faster convergence and higher accuracy, it suffers from pronounced fluctuations, whereas model aggregation offers greater stability but slower convergence and suboptimal accuracy. This paper presents FedQS, the first framework to theoretically analyze and address these disparities in SAFL. FedQS introduces a divide-and-conquer strategy to handle client heterogeneity by classifying clients into four distinct types and adaptively optimizing their local training based on data distribution characteristics and available computational resources. Extensive experiments on computer vision, natural language processing, and real-world tasks demonstrate that FedQS achieves the highest accuracy, attains the lowest loss, and ranks among the fastest in convergence speed, outperforming state-of-the-art baselines. Our work bridges the gap between aggregation strategies in SAFL, offering a unified solution for stable, accurate, and efficient federated learning. The code and datasets are available at https://github.com/bkjod/FedQS_.

cs.LG↗

arXiv2025

HunyuanVideo 1.5 Technical Report

We present HunyuanVideo 1.5, a lightweight yet powerful open-source video generation model that achieves state-of-the-art visual quality and motion coherence with only 8.3 billion parameters, enabling efficient inference on consumer-grade GPUs. This achievement is built upon several key components, including meticulous data curation, an advanced DiT architecture featuring selective and sliding tile attention (SSTA), enhanced bilingual understanding through glyph-aware text encoding, progressive pre-training and post-training, and an efficient video super-resolution network. Leveraging these designs, we developed a unified framework capable of high-quality text-to-video and image-to-video generation across multiple durations and resolutions. Extensive experiments demonstrate that this compact and proficient model establishes a new state-of-the-art among open-source video generation models. By releasing the code and model weights, we provide the community with a high-performance foundation that lowers the barrier to video creation and research, making advanced video generation accessible to a broader audience. All open-source assets are publicly available at https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5.

cs.CV↗

arXiv2025

Use of Continuous Glucose Monitoring with Machine Learning to Identify Metabolic Subphenotypes and Inform Precision Lifestyle Changes

The classification of diabetes and prediabetes by static glucose thresholds obscures the pathophysiological dysglycemia heterogeneity, primarily driven by insulin resistance (IR), beta-cell dysfunction, and incretin deficiency. This review demonstrates that continuous glucose monitoring and wearable technologies enable a paradigm shift towards non-invasive, dynamic metabolic phenotyping. We show evidence that machine learning models can leverage high-resolution glucose data from at-home, CGM-enabled oral glucose tolerance tests to accurately predict gold-standard measures of muscle IR and beta-cell function. This personalized characterization extends to real-world nutrition, where an individual's unique postprandial glycemic response (PPGR) to standardized meals, such as the relative glucose spike to potatoes versus grapes, could serve as a biomarker for their metabolic subtype. Moreover, integrating wearable data reveals that habitual diet, sleep, and physical activity patterns, particularly their timing, are uniquely associated with specific metabolic dysfunctions, informing precision lifestyle interventions. The efficacy of dietary mitigators in attenuating PPGR is also shown to be phenotype-dependent. Collectively, this evidence demonstrates that CGM can deconstruct the complexity of early dysglycemia into distinct, actionable subphenotypes. This approach moves beyond simple glycemic control, paving the way for targeted nutritional, behavioral, and pharmacological strategies tailored to an individual's core metabolic defects, thereby paving the way for a new era of precision diabetes prevention.

cs.LG↗

arXiv2025

KAN-Enhanced Contrastive Learning Accelerating Crystal Structure Identification from XRD Patterns

Accurate determination of crystal structures is central to materials science, underpinning the understanding of composition-structure-property relationships and the discovery of new materials. Powder X-ray diffraction is a key technique in this pursuit due to its versatility and reliability. However, current analysis pipelines still rely heavily on expert knowledge and slow iterative fitting, limiting their scalability in high-throughput and autonomous settings. Here, we introduce a physics-guided contrastive learning framework termed as XCCP. It aligns powder diffraction patterns with candidate crystal structures in a shared embedding space to enable efficient structure retrieval and symmetry recognition. The XRD encoder employs a dual-expert design with a Kolmogorov-Arnold Network projection head, one branch emphasizes low angle reflections reflecting long-range order, while the other captures dense high angle peaks shaped by symmetry. Coupled with a crystal graph encoder, contrastive pretraining yields physically grounded representations. XCCP demonstrates strong performance across tasks, with structure retrieval reaching 0.89 and space group identification attains 0.93 accuracy. The framework further generalizes to compositionally similar multi principal element alloys and demonstrates zero-shot transfer to experimental patterns. These results establish XCCP as a robust, interpretable, and scalable approach that offers a new paradigm for X-ray diffraction analysis. XCCP facilitates high-throughput screening, rapid structural validation and integration into autonomous laboratories.

cond-mat.mtrl-sci↗

arXiv2025

Impact of Wave Interference on the Consistency Relations of Internal Gravity Waves near the Ocean Bottom

Consistency relations of internal gravity waves (IGWs) describe ratios of cross-spectral quantities as functions of frequency. It has been a common practice to evaluate the measured or simulated signals (e.g., time series of velocity, density, etc.) against the consistency relations, as a way to determine whether an oceanic field of interest is comprised of IGWs. One such study is carried out in Nelson et al. (JGR Oceans, 125(5), 2020, e2019JC015974), which certifies that the ocean interior field in a numerical simulation of a region southwest of Hawaii is dominated by IGWs, through evaluating the consistency relations derived from time series at a depth of 620 m. However, we find that when the same procedure is applied at greater depths (e.g., 2362 m, 3062 m, and 4987 m), a clear deviation of the simulated signal from the classical consistency relations is observed. In this paper, we identify the reason for the unexpected deviation and show that it is a general phenomenon due to interference of low vertical modes under the reflection by the ocean bottom. We further derive a new set of formulae to characterize the consistency relations of these low modes and validate these formulae using model output.

physics.ao-ph↗

arXiv2025

Mechanism-Guided Residual Lifting and Control Consistent Modeling for Pneumatic Drying Processes

Pneumatic drying processes in industries such as agriculture, chemicals,and pharmaceuticals are notoriously difficult to model and control due to multi-source disturbances,coupled stage dynamics, and significant measurement delays. Traditional modeling paradigms often fail to simultaneously deliver accuracy, interpretability, and closed-loop applicability. To address this challenge, this paper introduces a unified hybrid modeling framework, termed Physics-Guided Residual Lifting with Control-Consistent Correction,which integrates a transient mechanistic model with a stability-constrained data-driven component. The framework covers the complete process chain of drying, transport, and winnowing. On the mechanistic level, the model unifies mass transfer dynamics using the partial pressure difference of water vapor, incorporates water activity clamping and latent heat corrections for bound water, and ensures energy closure with moisture-dependent specific heat. On the data-driven level,we propose an orthogonal residual learning scheme. It leverages intermediate states from the mechanistic model as proxy variables to construct a physics-inspired dictionary, preventing parameter compensation and overfitting during ridge regression. Furthermore, to ensure suitability for predictive control, a Control-Consistent Extended Dynamic Mode Decomposition with stability constraints is employed to learn the residual dynamics, for which we provide boundedness proofs and stability guarantees. The framework was validated on 10 industrial batches, comprising 63,000 samples. On unseen test data, the hybrid model achieved a Mean Absolute Error of 0.016% for outlet moisture and 0.015 °C for outlet temperature, with values improving to 0.986 and 0.995, respectively. The resulting prediction residuals exhibit white-noise characteristics, with significantly reduced spectral energy at low frequencies.

eess.SY↗

arXiv2025

Knowledge-based Visual Question Answer with Multimodal Processing, Retrieval and Filtering

Knowledge-based visual question answering (KB-VQA) requires visual language models (VLMs) to integrate visual understanding with external knowledge retrieval. Although retrieval-augmented generation (RAG) achieves significant advances in this task by combining knowledge-base querying, it still struggles with the quality of multimodal queries and the relevance of retrieved results. To overcome these challenges, we propose a novel three-stage method, termed Wiki-PRF, including Processing, Retrieval and Filtering stages. The processing stage dynamically invokes visual tools to extract precise multimodal information for retrieval. The retrieval stage integrates visual and text features to achieve multimodal knowledge retrieval. The filtering stage performs relevance filtering and concentration on retrieval results. To this end, we introduce a visual language model trained with answer accuracy and format consistency as reward signals via a reinforcement learning manner. This enhances the model's reasoning, tool invocation for accurate queries, and filtering of irrelevant content. Experiments on benchmark datasets (E-VQA and InfoSeek) show significant improvements~(36.0 and 42.8) in answer quality, achieving state-of-the-art performance. Code is available at https://github.com/cqu-student/Wiki-PRF

cs.CV↗

arXiv2025

RHINO: Guided Reasoning for Mapping Network Logs to Adversarial Tactics and Techniques with Large Language Models

Modern Network Intrusion Detection Systems generate vast volumes of low-level alerts, yet these outputs remain semantically fragmented, requiring labor-intensive manual correlation with high-level adversarial behaviors. Existing solutions for automating this mapping-rule-based systems and machine learning classifiers-suffer from critical limitations: rule-based approaches fail to adapt to novel attack variations, while machine learning methods lack contextual awareness and treat tactic-technique mapping as a syntactic matching problem rather than a reasoning task. Although Large Language Models have shown promise in cybersecurity tasks, preliminary experiments reveal that existing LLM-based methods frequently hallucinate technique names or produce decontextualized mappings due to their single-step classification approach. To address these challenges, we introduce RHINO, a novel framework that decomposes LLM-based attack analysis into three interpretable phases mirroring human reasoning: (1) behavioral abstraction, where raw logs are translated into contextualized narratives; (2) multi-role collaborative inference, generating candidate techniques by evaluating behavioral evidence against MITRE ATT&CK knowledge; and (3) validation, cross-referencing predictions with official MITRE definitions to rectify hallucinations. RHINO bridges the semantic gap between low-level observations and adversarial intent while improving output reliability through structured reasoning. We evaluate RHINO on three benchmarks across four backbone models. RHINO achieved high accuracy, with model performance ranging from 86.38% to 88.45%, resulting in relative gains from 24.25% to 76.50% across different models. Our results demonstrate that RHINO significantly enhances the interpretability and scalability of threat analysis, offering a blueprint for deploying LLMs in operational security settings.

cs.CR↗

arXiv2025

Some Unified Theory for Variance Reduced Prox-Linear Methods

This work considers the nonconvex, nonsmooth problem of minimizing a composite objective of the form $f(g(x))+h(x)$ where the inner mapping $g$ is a smooth finite summation or expectation amenable to variance reduction. In such settings, prox-linear methods can enjoy variance-reduced speed-ups despite the existence of nonsmoothness. We provide a unified convergence theory applicable to a wide range of common variance-reduced vector and Jacobian constructions. All the technical conditions we required for variance-reduced methods can be summarized in a single unified assumption. Our theory (i) only requires operator norm bounds on Jacobians (whereas prior works used potentially much larger Frobenius norms), (ii) provides state-of-the-art high probability guarantees, and (iii) allows inexactness in proximal computations.

math.OC↗

arXiv2025

On the Role of Preference Variance in Preference Optimization

Direct Preference Optimization (DPO) has emerged as an important approach for learning from human preferences in aligning large language models (LLMs). However, collecting human preference data is costly and inefficient, motivating methods to reduce the required annotations. In this work, we investigate the impact of \emph{preference variance} (PVar), which measures the variance in model preferences when comparing pairs of responses, on the effectiveness of DPO training. We provide a theoretical insight by establishing an upper bound on the DPO gradient norm for any given prompt, showing it is controlled by the PVar of that prompt. This implies that prompts with low PVar can only produce small gradient updates, making them less valuable for learning. We validate this finding by fine-tuning LLMs with preferences generated by a reward model, evaluating on two benchmarks (AlpacaEval 2.0 and Arena-Hard). Experimental results demonstrate that prompts with higher PVar outperform randomly selected prompts or those with lower PVar. We also show that our PVar-based selection method is robust, when using smaller reward models (1B, 3B) for selection. Notably, in a separate experiment using the original human annotations from the UltraFeedback dataset, we found that training on only the top 10\% of prompts with the highest PVar yields better evaluation performance than training on the full dataset, highlighting the importance of preference variance in identifying informative examples for efficient LLM alignment.

cs.CL↗

arXiv2025

ExpVid: A Benchmark for Experiment Video Understanding & Reasoning

Multimodal Large Language Models (MLLMs) hold promise for accelerating scientific discovery by interpreting complex experimental procedures. However, their true capabilities are poorly understood, as existing benchmarks neglect the fine-grained and long-horizon nature of authentic laboratory work, especially in wet-lab settings. To bridge this gap, we introduce ExpVid, the first benchmark designed to systematically evaluate MLLMs on scientific experiment videos. Curated from peer-reviewed video publications, ExpVid features a new three-level task hierarchy that mirrors the scientific process: (1) Fine-grained Perception of tools, materials, and actions; (2) Procedural Understanding of step order and completeness; and (3) Scientific Reasoning that connects the full experiment to its published conclusions. Our vision-centric annotation pipeline, combining automated generation with multi-disciplinary expert validation, ensures that tasks require visual grounding. We evaluate 19 leading MLLMs on ExpVid and find that while they excel at coarse-grained recognition, they struggle with disambiguating fine details, tracking state changes over time, and linking experimental procedures to scientific outcomes. Our results reveal a notable performance gap between proprietary and open-source models, particularly in high-order reasoning. ExpVid not only provides a diagnostic tool but also charts a roadmap for developing MLLMs capable of becoming trustworthy partners in scientific experimentation.

cs.CV↗

arXiv2025

Flora: Effortless Context Construction to Arbitrary Length and Scale

Effectively handling long contexts is challenging for Large Language Models (LLMs) due to the rarity of long texts, high computational demands, and substantial forgetting of short-context abilities. Recent approaches have attempted to construct long contexts for instruction tuning, but these methods often require LLMs or human interventions, which are both costly and limited in length and diversity. Also, the drop in short-context performances of present long-context LLMs remains significant. In this paper, we introduce Flora, an effortless (human/LLM-free) long-context construction strategy. Flora can markedly enhance the long-context performance of LLMs by arbitrarily assembling short instructions based on categories and instructing LLMs to generate responses based on long-context meta-instructions. This enables Flora to produce contexts of arbitrary length and scale with rich diversity, while only slightly compromising short-context performance. Experiments on Llama3-8B-Instruct and QwQ-32B show that LLMs enhanced by Flora excel in three long-context benchmarks while maintaining strong performances in short-context tasks. Our data-construction code is available at \href{https://github.com/txchen-USTC/Flora}{https://github.com/txchen-USTC/Flora}.

cs.CL↗

arXiv2025

TALENT: Table VQA via Augmented Language-Enhanced Natural-text Transcription

Table Visual Question Answering (Table VQA) is typically addressed by large vision-language models (VLMs). While such models can answer directly from images, they often miss fine-grained details unless scaled to very large sizes, which are computationally prohibitive, especially for mobile deployment. A lighter alternative is to have a small VLM perform OCR and then use a large language model (LLM) to reason over structured outputs such as Markdown tables. However, these representations are not naturally optimized for LLMs and still introduce substantial errors. We propose TALENT (Table VQA via Augmented Language-Enhanced Natural-text Transcription), a lightweight framework that leverages dual representations of tables. TALENT prompts a small VLM to produce both OCR text and natural language narration, then combines them with the question for reasoning by an LLM. This reframes Table VQA as an LLM-centric multimodal reasoning task, where the VLM serves as a perception-narration module rather than a monolithic solver. Additionally, we construct ReTabVQA, a more challenging Table VQA dataset requiring multi-step quantitative reasoning over table images. Experiments show that TALENT enables a small VLM-LLM combination to match or surpass a single large VLM at significantly lower computational cost on both public datasets and ReTabVQA.

cs.CL↗

arXiv2025

Interference Alignment for Multi-cluster Over-the-Air Computation

One of the main challenges facing Internet of Things (IoT) networks is managing interference caused by the large number of devices communicating simultaneously, particularly in multi-cluster networks where multiple devices simultaneously transmit to their respective receiver. Over-the-Air Computation (AirComp) has emerged as a promising solution for efficient real-time data aggregation, yet its performance suffers in dense, interference-limited environments. To address this, we propose a novel Interference Alignment (IA) scheme tailored for up-link AirComp systems. Unlike previous approaches, the proposed method scales to an arbitrary number $\sf K$ of clusters and enables each cluster to exploit half of the available channels, instead of only $\tfrac{1}{\sf K}$ as in time-sharing. In addition, we develop schemes tailored to scenarios where users are shared between adjacent clusters.

eess.SP↗

arXiv2025

Minimum-Weight Parity Factor Decoder for Quantum Error Correction

Fast and accurate quantum error correction (QEC) decoding is crucial for scalable fault-tolerant quantum computation. Most-Likely-Error (MLE) decoding, while being near-optimal, is intractable on general quantum Low-Density Parity-Check (qLDPC) codes and typically relies on approximation and heuristics. We propose HyperBlossom, a unified framework that formulates MLE decoding as a Minimum-Weight Parity Factor (MWPF) problem and generalizes the blossom algorithm to hypergraphs via a similar primal-dual linear programming model with certifiable proximity bounds. HyperBlossom unifies all the existing graph-based decoders like (Hypergraph) Union-Find decoders and Minimum-Weight Perfect Matching (MWPM) decoder, thus bridging the gap between heuristic and certifying decoders. We implement HyperBlossom in software, namely Hyperion. Hyperion achieves a 4.8x lower logical error rate compared to the MWPM decoder on the distance-11 surface code and 1.6x lower logical error rate compared to a fine-tuned BPOSD decoder on the $[[90, 8, 10]]$ bivariate bicycle code under code-capacity noise. It also achieves an almost-linear average runtime scaling on both the surface code and the color code, with numerical results up to sufficiently large code distances of 99 and 31 for code-capacity noise and circuit-level noise, respectively.

quant-ph↗

arXiv2025

SD-VLM: Spatial Measuring and Understanding with Depth-Encoded Vision-Language Models

While vision language models (VLMs) excel in 2D semantic visual understanding, their ability to quantitatively reason about 3D spatial relationships remains under-explored, due to the deficiency of 2D images' spatial representation ability. In this paper, we analyze the problem hindering VLMs' spatial understanding abilities and propose SD-VLM, a novel framework that significantly enhances fundamental spatial perception abilities of VLMs through two key contributions: (1) propose Massive Spatial Measuring and Understanding (MSMU) dataset with precise spatial annotations, and (2) introduce a simple depth positional encoding method strengthening VLMs' spatial awareness. MSMU dataset covers massive quantitative spatial tasks with 700K QA pairs, 2.5M physical numerical annotations, and 10K chain-of-thought augmented samples. We have trained SD-VLM, a strong generalist VLM which shows superior quantitative spatial measuring and understanding capability. SD-VLM not only achieves state-of-the-art performance on our proposed MSMU-Bench, but also shows spatial generalization abilities on other spatial understanding benchmarks including Q-Spatial and SpatialRGPT-Bench. Extensive experiments demonstrate that SD-VLM outperforms GPT-4o and Intern-VL3-78B by 26.91% and 25.56% respectively on MSMU-Bench. Code and models are released at https://github.com/cpystan/SD-VLM.

cs.CV↗

arXiv2025

A comprehensive study of time delay between optical/near-infrared and X-ray emissions in black hole X-ray binaries

We conducted a comprehensive study of daily delays using multi-wavelength data from a sample of well-studied black hole X-ray binaries, specifically focusing on the sources GX 339-4, 4U 1543-47, and XTE J1550-564. The Interpolated Cross-Correlation Function method was employed to investigate the temporal relationship between the X-ray (Compton component) and optical-infrared (OIR) emissions. Our results show that during the rising hard state, the Compton emission consistently lags behind OIR emission for several days. In contrast, during the decaying hard state, the OIR emission lags behind the Compton emission by approximately 6 to 35 days. This measurement can potentially be used in models of accretion physics and disk instability. We explore the underlying mechanisms responsible for these time delays, highlighting the critical role of viscous heating in the accretion disk in generating OIR luminosity for these sources. The observed time delays during both the rising and decaying hard states are well explained by the disk instability model.

astro-ph.HE↗