arXiv subjects

Xiao Han

Publications and source records attributed to Xiao Han.

At least 73 records · Page 4Linked to original sources

arXiv2026

A Nonlocal Orientation Field Phase-Field Model for Misorientation- and Inclination- Dependent Grain Boundaries

We propose to incorporate grain boundary (GB) anisotropy in phase-field modeling by extending the standard partial differential equations formulation to include a nonlocal functional of an orientation field. Regardless of the number of grains in the simulation, the model uses a single orientation field and incorporates grain misorientation and inclination information obtained from sampling the orientation field at optimized locations in the vicinity of the grain boundary. The formalism enables simple and precise tuning of GB energy anisotropy while reducing an extensive fitting procedure. The functional includes an explicit GB anisotropy function to control the GB energy as a function of both misorientation and inclination. The model is validated by reproducing the linear grain growth rate, Wulff shapes with varying misorientations and anisotropic coefficients, and analytical equilibrium dihedral angles at triple junctions. Polycrystalline simulations demonstrate grain growth, coalescence, triple junction behavior, and the influence of anisotropy on grain morphology.

cond-mat.mtrl-sci↗

arXiv2026

R2LED: Equipping Retrieval and Refinement in Lifelong User Modeling with Semantic IDs for CTR Prediction

Lifelong user modeling, which leverages users' long-term behavior sequences for CTR prediction, has been widely applied in personalized services. Existing methods generally adopted a two-stage "retrieval-refinement" strategy to balance effectiveness and efficiency. However, they still suffer from (i) noisy retrieval due to skewed data distribution and (ii) lack of semantic understanding in refinement. While semantic enhancement, e.g., LLMs modeling or semantic embeddings, offers potential solutions to these two challenges, these approaches face impractical inference costs or insufficient representation granularity. Obsorbing multi-granularity and lightness merits of semantic identity (SID), we propose a novel paradigm that equips retrieval and refinement in Lifelong User Modeling with SEmantic IDs (R2LED) to address these issues. First, we introduce a Multi-route Mixed Retrieval for the retrieval stage. On the one hand, it captures users' interests from various granularities by several parallel recall routes. On the other hand, a mixed retrieval mechanism is proposed to efficiently retrieve candidates from both collaborative and semantic views, reducing noise. Then, for refinement, we design a Bi-level Fusion Refinement, including a target-aware cross-attention for route-level fusion and a gate mechanism for SID-level fusion. It can bridge the gap between semantic and collaborative spaces, exerting the merits of SID. The comprehensive experimental results on two public datasets demonstrate the superiority of our method in both performance and efficiency. To facilitate the reproduction, we have released the code online https://github.com/abananbao/R2LED.

cs.IR↗

arXiv2026

Enhancing Conversational Agents via Task-Oriented Adversarial Memory Adaptation

Conversational agents struggle to handle long conversations due to context window limitations. Therefore, memory systems are developed to leverage essential historical information. Existing memory systems typically follow a pipeline of offline memory construction and update, and online retrieval. Despite the flexible online phase, the offline phase remains fixed and task-independent. In this phase, memory construction operates under a predefined workflow and fails to emphasize task relevant information. Meanwhile, memory updates are guided by generic metrics rather than task specific supervision. This leads to a misalignment between offline memory preparation and task requirements, which undermines downstream task performance. To this end, we propose an Adversarial Memory Adaptation mechanism (AMA) that aligns memory construction and update with task objectives by simulating task execution. Specifically, first, a challenger agent generates question answer pairs based on the original dialogues. The constructed memory is then used to answer these questions, simulating downstream inference. Subsequently, an evaluator agent assesses the responses and performs error analysis. Finally, an adapter agent analyzes the error cases and performs dual level updates on both the construction strategy and the content. Through this process, the memory system receives task aware supervision signals in advance during the offline phase, enhancing its adaptability to downstream tasks. AMA can be integrated into various existing memory systems, and extensive experiments on long dialogue benchmark LoCoMo demonstrate its effectiveness.

cs.CL↗

arXiv2026

Multimodal Trajectory Representation Learning for Travel Time Estimation

Accurate travel time estimation (TTE) plays a crucial role in intelligent transportation systems. However, it remains challenging due to heterogeneous data sources and complex traffic dynamics. Moreover, traditional approaches typically convert trajectory data into fixed-length representations. This overlooks the inherent variability of real-world motion patterns, often resulting in information loss and redundancy. To address these challenges, this paper introduces the Multimodal Dynamic Trajectory Integration (MDTI) framework--a novel multimodal trajectory representation learning approach that integrates GPS sequences, grid trajectories, and road network constraints to enhance the performance of TTE. MDTI employs modality-specific encoders and a multimodal fusion module to capture complementary spatial, temporal, and topological semantics, while a dynamic trajectory modeling mechanism adaptively regulates information density for trajectories of varying lengths. Two self-supervised pretraining objectives, named contrastive alignment and masked language modeling, further strengthen multimodal consistency and contextual understanding. Extensive experiments on three real-world datasets demonstrate that MDTI consistently outperforms state-of-the-art baselines, confirming its robustness and strong generalization abilities. The code is publicly available at: https://github.com/City-Computing/MDTI.

cs.LG↗

arXiv2025

Degree-Weighted Social Learning

We study social learning in which agents weight neighbors' opinions differently based on their degrees, capturing situations in which agents place more trust in well-connected individuals or, conversely, discount their influence. We derive asymptotic properties of learning outcomes in large stochastic networks and analyze how the weighting rule affects societal wisdom and convergence speed. We find that assigning greater weight to higher-degree neighbors harms wisdom but has a non-monotonic effect on convergence speed, depending on the diversity of views within high- and low-degree groups, highlighting a potential trade-off between convergence speed and wisdom.

econ.TH↗

arXiv2025

Quantum jet Hopf algebroids by cotwist

We introduce a cotwist construction for Hopf algebroids that also entails cotwisting or `quantisation' of the base and which is dual to a previous twisting construction of P. Xu. Whereas the latter applied the construction to the algebra of differential operators on a classical base $B$, we show that the dual of this is the algebra of sections $J(B)$ of the jet bundle and hence that the latter forms a Hopf algebroid. This is constructed for commutative algebras $B$ in a pro-object setting via quotients of the pair Hopf algebroid $B\otimes B$ and can then be deformed by our cotwist construction to give a possibly noncommutative jet Hopf algebroid over a noncommutative base. We also observe in the commutative case that $J^k(B)$ for jets of order $k$ can be identified with $J^1(B_k)$ where $B_k$ denotes $B$ equipped with a certain noncommutative first order differential calculus.

math.QA↗

arXiv2025

A few observations around Gaussian domination and continuous symmetry breaking for spin O(N) model

We investigate the notion of Gaussian domination for the spin $O(N)$ model on general finite graphs. We begin by proving a general inequality for spin correlations under the assumption of Gaussian domination, which directly implies long-range order at low temperatures for graphs with bounded Green's function. Usually, Gaussian domination is proved via reflection positivity, but this requires strict symmetries and is very rigid. In this article we also probe the boundaries of elementary methods for proving Gaussian domination. Although we did not find a way to get uniform bounds, we do offer new views for Gaussian domination at low and high temperatures for finite graphs, and a few counterexamples illustrating the interplay between correlation estimates and Gaussian domination and how local changes in the graph structure can affect the presence of Gaussian domination.

math.PR↗

arXiv2025

A new bound for the Fourier-Entropy-Influence conjecture

In this paper, we prove that the Fourier entropy of an $n$-dimensional boolean function $f$ can be upper-bounded by $O(I(f)+ \sum\limits_{k\in[n]}I_k(f)\log \frac{1}{I_k(f)})$, where $I(f)$ is its total influence and $I_k(f)$ is the influence of the $k$-th coordinate. The proof is elementary and uses iterative bounds on moments of Fourier coefficients over different levels.

math.CO↗

arXiv2025

Scaling Zero-Shot Reference-to-Video Generation

Reference-to-video (R2V) generation aims to synthesize videos that align with a text prompt while preserving the subject identity from reference images. However, current R2V methods are hindered by the reliance on explicit reference image-video-text triplets, whose construction is highly expensive and difficult to scale. We bypass this bottleneck by introducing Saber, a scalable zero-shot framework that requires no explicit R2V data. Trained exclusively on video-text pairs, Saber employs a masked training strategy and a tailored attention-based model design to learn identity-consistent and reference-aware representations. Mask augmentation techniques are further integrated to mitigate copy-paste artifacts common in reference-to-video generation. Moreover, Saber demonstrates remarkable generalization capabilities across a varying number of references and achieves superior performance on the OpenS2V-Eval benchmark compared to methods trained with R2V data.

cs.CV↗

arXiv2025

Generalized Linear Spectral Statistics of High-dimensional Sample Covariance Matrices and Its Applications

In this paper, we introduce the \textbf{G}eneralized \textbf{L}inear \textbf{S}pectral \textbf{S}tatistics (GLSS) of a high-dimensional sample covariance matrix $\bm{S}_n$, denoted as $\operatorname{tr}f(\bm{S}_n)\bm{B}_n$, which effectively captures distinct spectral properties of $\bm{S}_n$ by incorporating an ancillary matrix $\bm{B}_n$ and a test function $f$. The joint asymptotic normality of GLSS associated with different test functions is established under mild assumptions on $\bm{B}_n$ and the underlying distribution, when the dimension $n$ and sample size $N$ are comparable. The convergence rate of GLSS is determined by $\sqrt{{N}/{\operatorname{rank}(\bm{B}_n)}}$. Subsequently, we propose a novel functional projection approach based on GLSS for hypothesis testing on eigenspaces of ``population-spiked'' covariance matrices, showcasing a universality phenomenon in the magnitude of the spikes. The theoretical accuracy of our results established for GLSS and the advantages of the newly suggested testing procedure are demonstrated through various numerical studies.

math.ST↗

arXiv2025

TUNA: Taming Unified Visual Representations for Native Unified Multimodal Models

Unified multimodal models (UMMs) aim to jointly perform multimodal understanding and generation within a single framework. We present TUNA, a native UMM that builds a unified continuous visual representation by cascading a VAE encoder with a representation encoder. This unified representation space allows end-to-end processing of images and videos for both understanding and generation tasks. Compared to prior UMMs with decoupled representations, TUNA's unified visual space avoids representation format mismatches introduced by separate encoders, outperforming decoupled alternatives in both understanding and generation. Moreover, we observe that stronger pretrained representation encoders consistently yield better performance across all multimodal tasks, highlighting the importance of the representation encoder. Finally, in this unified setting, jointly training on both understanding and generation data allows the two tasks to benefit from each other rather than interfere. Our extensive experiments on multimodal understanding and generation benchmarks show that TUNA achieves state-of-the-art results in image and video understanding, image and video generation, and image editing, demonstrating the effectiveness and scalability of its unified representation design.

cs.CV↗

arXiv2025

Data Efficient Adaptation in Large Language Models via Continuous Low-Rank Fine-Tuning

Recent advancements in Large Language Models (LLMs) have emphasized the critical role of fine-tuning (FT) techniques in adapting LLMs to specific tasks, especially when retraining from scratch is computationally infeasible. Fine-tuning enables LLMs to leverage task- or domain-specific data, producing models that more effectively meet the requirements of targeted applications. However, conventional FT approaches often suffer from catastrophic forgetting and suboptimal data efficiency, limiting their real-world applicability. To address these challenges, this paper proposes \textbf{DEAL}, a novel framework that integrates Low-Rank Adaptation (LoRA) with a continuous fine-tuning strategy. By incorporating knowledge retention and adaptive parameter update modules, the framework mitigates the limitations of existing FT methods while maintaining efficiency. Experiments on 15 diverse datasets show that DEAL consistently outperforms baseline methods, yielding substantial gains in task accuracy and resource efficiency. These findings demonstrate the potential of our approach to advance continual adaptation in LLMs by enhancing task performance while improving resource efficiency. The source code is publicly available at https://github.com/zzm-black/DEAL-Continuous-Low-Rank-Fine-Tuning.

cs.AI↗

arXiv2025

Contextual Attention Modulation: Towards Efficient Multi-Task Adaptation in Large Language Models

Large Language Models (LLMs) possess remarkable generalization capabilities but struggle with multi-task adaptation, particularly in balancing knowledge retention with task-specific specialization. Conventional fine-tuning methods suffer from catastrophic forgetting and substantial resource consumption, while existing parameter-efficient methods perform suboptimally in complex multi-task scenarios. To address this, we propose Contextual Attention Modulation (CAM), a novel mechanism that dynamically modulates the representations of self-attention modules in LLMs. CAM enhances task-specific features while preserving general knowledge, thereby facilitating more effective and efficient adaptation. For effective multi-task adaptation, CAM is integrated into our Hybrid Contextual Attention Modulation (HyCAM) framework, which combines a shared, full-parameter CAM module with multiple specialized, lightweight CAM modules, enhanced by a dynamic routing strategy for adaptive knowledge fusion. Extensive experiments on heterogeneous tasks, including question answering, code generation, and logical reasoning, demonstrate that our approach significantly outperforms existing approaches, achieving an average performance improvement of 3.65%. The implemented code and data are available to ease reproducibility at https://github.com/Applied-Machine-Learning-Lab/HyCAM.

cs.AI↗

arXiv2025

PromoGuardian: Detecting Promotion Abuse Fraud with Multi-Relation Fused Graph Neural Networks

As e-commerce platforms develop, fraudulent activities are increasingly emerging, posing significant threats to the security and stability of these platforms. Promotion abuse is one of the fastest-growing types of fraud in recent years and is characterized by users exploiting promotional activities to gain financial benefits from the platform. To investigate this issue, we conduct the first study on promotion abuse fraud in e-commerce platforms MEITUAN. We find that promotion abuse fraud is a group-based fraudulent activity with two types of fraudulent activities: Stocking Up and Cashback Abuse. Unlike traditional fraudulent activities such as fake reviews, promotion abuse fraud typically involves ordinary customers conducting legitimate transactions and these two types of fraudulent activities are often intertwined. To address this issue, we propose leveraging additional information from the spatial and temporal perspectives to detect promotion abuse fraud. In this paper, we introduce PROMOGUARDIAN, a novel multi-relation fused graph neural network that integrates the spatial and temporal information of transaction data into a homogeneous graph to detect promotion abuse fraud. We conduct extensive experiments on real-world data from MEITUAN, and the results demonstrate that our proposed model outperforms state-of-the-art methods in promotion abuse fraud detection, achieving 93.15% precision, detecting 2.1 to 5.0 times more fraudsters, and preventing 1.5 to 8.8 times more financial losses in production environments.

cs.CR↗

arXiv2025

Think as a Doctor: An Interpretable AI Approach for ICU Mortality Prediction

Intensive Care Unit (ICU) mortality prediction, which estimates a patient's mortality status at discharge using EHRs collected early in an ICU admission, is vital in critical care. For this task, predictive accuracy alone is insufficient; interpretability is equally essential for building clinical trust and meeting regulatory standards, a topic that has attracted significant attention in information system research. Accordingly, an ideal solution should enable intrinsic interpretability and align its reasoning with three key elements of the ICU decision-making practices: clinical course identification, demographic heterogeneity, and prognostication awareness. However, conventional approaches largely focus on demographic heterogeneity, overlooking clinical course identification and prognostication awareness. Recent prototype learning methods address clinical course identification, yet the integration of the other elements into such frameworks remains underexplored. To address these gaps, we propose ProtoDoctor, a novel ICU mortality prediction framework that delivers intrinsic interpretability while integrating all three elements of the ICU decision-making practices into its reasoning process. Methodologically, ProtoDoctor features two key innovations: the Prognostic Clinical Course Identification module and the Demographic Heterogeneity Recognition module. The former enables the identification of clinical courses via prototype learning and achieves prognostication awareness using a novel regularization mechanism. The latter models demographic heterogeneity through cohort-specific prototypes and risk adjustments. Extensive empirical evaluations demonstrate that ProtoDoctor outperforms state-of-the-art baselines in predictive accuracy. Human evaluations further confirm that its interpretations are more clinically meaningful, trustworthy, and applicable in ICU practice.

cs.LG↗

arXiv2025

Enhancing Diffusion Policy with Classifier-Free Guidance for Temporal Robotic Tasks

Temporal sequential tasks challenge humanoid robots, as existing Diffusion Policy (DP) and Action Chunking with Transformers (ACT) methods often lack temporal context, resulting in local optima traps and excessive repetitive actions. To address these issues, this paper introduces a Classifier-Free Guidance-Based Diffusion Policy (CFG-DP), a novel framework to enhance DP by integrating Classifier-Free Guidance (CFG) with conditional and unconditional models. Specifically, CFG leverages timestep inputs to track task progression and ensure precise cycle termination. It dynamically adjusts action predictions based on task phase, using a guidance factor tuned to balance temporal coherence and action accuracy. Real-world experiments on a humanoid robot demonstrate high success rates and minimal repetitive actions. Furthermore, we assessed the model's ability to terminate actions and examined how different components and parameter adjustments affect its performance. This framework significantly enhances deterministic control and execution reliability for sequential robotic tasks.

cs.RO↗

arXiv2025

Algorithmic Recourse in Abnormal Multivariate Time Series

Algorithmic recourse provides actionable recommendations to alter unfavorable predictions of machine learning models, enhancing transparency through counterfactual explanations. While significant progress has been made in algorithmic recourse for static data, such as tabular and image data, limited research explores recourse for multivariate time series, particularly for reversing abnormal time series. This paper introduces Recourse in time series Anomaly Detection (RecAD), a framework for addressing anomalies in multivariate time series using backtracking counterfactual reasoning. By modeling the causes of anomalies as external interventions on exogenous variables, RecAD predicts recourse actions to restore normal status as counterfactual explanations, where the recourse function, responsible for generating actions based on observed data, is trained using an end-to-end approach. Experiments on synthetic and real-world datasets demonstrate its effectiveness.

cs.LG↗

arXiv2025

Learning-based Privacy-Preserving Graph Publishing Against Sensitive Link Inference Attacks

Publishing graph data is widely desired to enable a variety of structural analyses and downstream tasks. However, it also potentially poses severe privacy leakage, as attackers may leverage the released graph data to launch attacks and precisely infer private information such as the existence of hidden sensitive links in the graph. Prior studies on privacy-preserving graph data publishing relied on heuristic graph modification strategies and it is difficult to determine the graph with the optimal privacy--utility trade-off for publishing. In contrast, we propose the first privacy-preserving graph structure learning framework against sensitive link inference attacks, named PPGSL, which can automatically learn a graph with the optimal privacy--utility trade-off. The PPGSL operates by first simulating a powerful surrogate attacker conducting sensitive link attacks on a given graph. It then trains a parameterized graph to defend against the simulated adversarial attacks while maintaining the favorable utility of the original graph. To learn the parameters of both parts of the PPGSL, we introduce a secure iterative training protocol. It can enhance privacy preservation and ensure stable convergence during the training process, as supported by the theoretical proof. Additionally, we incorporate multiple acceleration techniques to improve the efficiency of the PPGSL in handling large-scale graphs. The experimental results confirm that the PPGSL achieves state-of-the-art privacy--utility trade-off performance and effectively thwarts various sensitive link inference attacks.

cs.CR↗