arXiv subjects

Lu Chen

Publications and source records attributed to Lu Chen.

At least 91 records · Page 5Linked to original sources

arXiv2026

Replacing Multi-Step Assembly of Data Preparation Pipelines with One-Step LLM Pipeline Generation for Table QA

Table Question Answering (TQA) aims to answer natural language questions over structured tables. Large Language Models (LLMs) enable promising solutions to this problem, with operator-centric solutions that generate table manipulation pipelines in a multi-step manner offering state-of-the-art performance. However, these solutions rely on multiple LLM calls, resulting in prohibitive latencies and computational costs. We propose Operation-R1, the first framework that trains lightweight LLMs (e.g., Qwen-4B/1.7B) via a novel variant of reinforcement learning with verifiable rewards to produce high-quality data-preparation pipelines for TQA in a single inference step. To train such an LLM, we first introduce a self-supervised rewarding mechanism to automatically obtain fine-grained pipeline-wise supervision signals for LLM training. We also propose variance-aware group resampling to mitigate training instability. To further enhance robustness of pipeline generation, we develop two complementary mechanisms: operation merge, which filters spurious operations through multi-candidate consensus, and adaptive rollback, which offers runtime protection against information loss in data transformation. Experiments on two benchmark datasets show that, with the same LLM backbone, Operation-R1 achieves average absolute accuracy gains of 8.83 and 4.44 percentage points over multi-step preparation baselines, with 79\% table compression and a 2.2$\times$ reduction in monetary cost.

cs.DB↗

arXiv2026

AirQA: A Comprehensive QA Dataset for AI Research with Instance-Level Evaluation

The growing volume of academic papers has made it increasingly difficult for researchers to efficiently extract key information. While large language models (LLMs) based agents are capable of automating question answering (QA) workflows for scientific papers, there still lacks a comprehensive and realistic benchmark to evaluate their capabilities. Moreover, training an interactive agent for this specific task is hindered by the shortage of high-quality interaction trajectories. In this work, we propose AirQA, a human-annotated comprehensive paper QA dataset in the field of artificial intelligence (AI), with 13,956 papers and 1,246 questions, that encompasses multi-task, multi-modal and instance-level evaluation. Furthermore, we propose ExTrActor, an automated framework for instruction data synthesis. With three LLM-based agents, ExTrActor can perform example generation and trajectory collection without human intervention. Evaluations of multiple open-source and proprietary models show that most models underperform on AirQA, demonstrating the quality of our dataset. Extensive experiments confirm that ExTrActor consistently improves the multi-turn tool-use capability of small models, enabling them to achieve performance comparable to larger ones.

cs.CL↗

arXiv2026

Sharp capillary Sobolev inequality and Moser-Trudinger inequality outside convex domain

The theory of sharp geometric inequality in $\mathbb{R}^n$ and inside convex cone has been well-developed, much less known for sharp capillary geometric inequality outside convex domain. Recently, Fusco-Julin-Morini-Pratelli \cite{FJMP} obtained sharp capillary isoperimetric inequality and make it possible to obtain the sharp capillary geometric inequality outside convex domain. In this paper, we establish the sharp capillary Sobolev inequality and Moser-Trudinger inequality outside convex domain, which can be seen as geometric inequality on the Finsler manifold to some extent. Our method is based on constructing capillary Pálya-Szegö rearrangement inequality outside convex domain. Finally, we also consider the capillary Talenti-Comparison principle and Bossel-Daners inequality.

math.AP↗

arXiv2026

Federated Learning and Unlearning for Recommendation with Personalized Data Sharing

Federated recommender systems (FedRS) have emerged as a paradigm for protecting user privacy by keeping interaction data on local devices while coordinating model training through a central server. However, most existing federated recommender systems adopt a one-size-fits-all assumption on user privacy, where all users are required to keep their data strictly local. This setting overlooks users who are willing to share their data with the server in exchange for better recommendation performance. Although several recent studies have explored personalized user data sharing in FedRS, they assume static user privacy preferences and cannot handle user requests to remove previously shared data and its corresponding influence on the trained model. To address this limitation, we propose FedShare, a federated learn-unlearn framework for recommender systems with personalized user data sharing. FedShare not only allows users to control how much interaction data is shared with the server, but also supports data unsharing requests by removing the influence of the unshared data from the trained model. Specifically, FedShare leverages shared data to construct a server-side high-order user-item graph and uses contrastive learning to jointly align local and global representations. In the unlearning phase, we design a contrastive unlearning mechanism that selectively removes representations induced by the unshared data using a small number of historical embedding snapshots, avoiding the need to store large amounts of historical gradient information as required by existing federated recommendation unlearning methods. Extensive experiments on three public datasets demonstrate that FedShare achieves strong recommendation performance in both the learning and unlearning phases, while significantly reducing storage overhead in the unlearning phase compared with state-of-the-art baselines.

cs.IR↗

arXiv2026

The Sommerfeld-Rellich Framework for Scattering on Hyperbolic Space: Far-Field Patterns and Inverse Problems

This paper establishes a complete time-harmonic scattering theory for hyperbolic space, formulating it within the classical Sommerfeld-Rellich paradigm centered on far-field patterns--a foundational framework that has been absent despite the well-developed spectral and time-dependent theories for this geometry. We explicitly construct the ingoing and outgoing fundamental solutions for the Helmholtz operator and perform a precise asymptotic analysis at the conformal boundary to derive a {hyperbolic Sommerfeld radiation condition}. This condition, which is a local criterion at infinity, uniquely selects physically admissible outgoing solutions. We prove a {hyperbolic Rellich theorem} guaranteeing the uniqueness of the scattered field and its far-field pattern from asymptotic data. Within this rigorous framework, we solve the direct scattering problem for compact sources, penetrable media, and impenetrable obstacles, providing explicit representations for the corresponding {far-field patterns}. As a principal application and demonstration of the framework's utility, we initiate the study of inverse scattering on hyperbolic space, formulating both the inverse obstacle and inverse medium problems where the objective is to reconstruct the scatterer from measurements of its far-field pattern. Our work lays the groundwork for a far-field-based approach to scattering and inversion in hyperbolic geometry. Due to the local feature of the {hyperbolic Sommerfeld radiation condition}, our study can be readily extended to the broader asymptotically hyperbolic manifolds.

math.AP↗

arXiv2026

DiSRouter: Distributed Self-Routing for LLM Selections

The proliferation of Large Language Models (LLMs) has created a diverse ecosystem of models with highly varying performance and costs, necessitating effective query routing to balance performance and expense. Current routing systems often rely on a centralized external router trained on a fixed set of LLMs, making them inflexible and prone to poor performance since the small router can not fully understand the knowledge boundaries of different LLMs. We introduce DiSRouter (Distributed Self-Router), a novel paradigm that shifts from centralized control to distributed routing. In DiSRouter, a query traverses a network of LLM agents, each independently deciding whether to answer or route to other agents based on its own self-awareness, its ability to judge its competence. This distributed design offers superior flexibility, scalability, and generalizability. To enable this, we propose a two-stage Self-Awareness Training pipeline that enhances each LLM's self-awareness. Extensive experiments demonstrate that DiSRouter significantly outperforms existing routing methods in utility across various scenarios, effectively distinguishes between easy and hard queries, and shows strong generalization to out-of-domain tasks. Our work validates that leveraging an LLM's intrinsic self-awareness is more effective than external assessment, paving the way for more modular and efficient multi-agent systems.

cs.CL↗

arXiv2026

Morphological Effects on Bacterial Brownian Motion: Validation of a Chiral Two-Body Model

We systematically investigate how flagellar morphology governs the stability of bacterial Brownian motion, evaluating the effectiveness of a simplified chiral two-body model. This model, which effectively captures the specific bacterial morphology and significantly reduces computational cost, is used for simulating bacterial Brownian motion. Our results demonstrate that the model accurately reproduces the Brownian motion of bacteria for contour lengths $Λ\ge5.0$~\si{μm}, helix radii $0.2\le R\le 0.5$~\si{μm}, and pitch angles $π/6\leθ\le2π/9$. We find that the translational and rotational velocities of bacteria depend linearly on the motor rotation rate, independent of dynamic viscosity. Increasing helix radius and contour length leads to more elongated trajectories and enhances their linearity. Furthermore, longer contour lengths improve the stability of the bacterial forward motion. Collectively, these findings demonstrate the essential role of flagella in stabilizing bacterial Brownian motion and confirm the effectiveness of the chiral two-body model for simulating this phenomenon.

physics.flu-dyn↗

arXiv2026

Effects of Near-Field Hydrodynamic Interactions on Bacterial Dynamics Near a Solid Surface

Near-field hydrodynamic interactions between bacteria and no-slip solid surfaces are the main mechanism underlying surface entrapment of bacteria. In this study, we employ a chiral two-body model to simulate bacterial dynamics near the surface. The simulation results show that as bacteria approach the surface, their translational velocities and diffusion coefficients decrease. Under the combination of near-field hydrodynamic interactions and DLVO forces, bacteria reach a stable fixed point in the phase plane and follow circular trajectories at this point. In particular, bacteria with left-handed helical flagella exhibit clockwise circular motion on the surface. During this process, as the stable height increases, the near-field hydrodynamic interactions weaken. Consequently, the translational velocity of the bacteria parallel to the surface increases while the rotational velocity perpendicular to the surface decreases, collectively increasing the radius of curvature. Ultimately, our findings demonstrate that near-field hydrodynamic interactions significantly prolong the surface residence time of bacteria. Additionally, smaller stable heights further amplify this effect, resulting in longer residence times and enhanced surface entrapment.

cond-mat.soft↗

arXiv2026

Quantitative properties of the Hardy-type mean field equation

In this paper, we consider the following Hardy-type mean field equation \[ \left\{ {\begin{array}{*{20}{c}} { - Δu-\frac{1}{(1-|x|^2)^2} u = λe^u}, & {\rm in} \ \ B_1,\\ {\ \ \ \ u = 0,} &\ {\rm on}\ \partial B_1, \end{array}} \right. \] \[\] where $λ>0$ is small and $B_1$ is the standard unit disc of $\mathbb{R}^2$. Applying the moving plane method of hyperbolic space and the accurate expansion of heat kernel on hyperbolic space, we establish the radial symmetry and Brezis-Merle lemma for solutions of Hardy-type mean field equation. Meanwhile, we also derive the quantitative results for solutions of Hardy-type mean field equation, which improves significantly the compactness results for classical mean-field equation obtained by Brezis-Merle and Li-Shafrir. Furthermore, applying the local Pohozaev identity from scaling, blow-up analysis and a contradiction argument, we prove that the solutions are unique when $λ$ is sufficiently close to 0.

math.AP↗

arXiv2026

Effective and Efficient Cross-City Traffic Knowledge Transfer: A Privacy-Preserving Perspective

Traffic prediction aims to forecast future traffic conditions using historical traffic data, serving a crucial role in urban computing and transportation management. While transfer learning and federated learning have been employed to address the scarcity of traffic data by transferring traffic knowledge from data-rich to data-scarce cities without traffic data exchange, existing approaches in Federated Traffic Knowledge Transfer (FTT) still face several critical challenges such as potential privacy leakage, cross-city data distribution discrepancies, and low data quality, hindering their practical application in real-world scenarios. To this end, we present FedTT, a novel privacy-aware and efficient federated learning framework for cross-city traffic knowledge transfer. Specifically, our proposed framework includes three key innovations: (i) a traffic view imputation method for missing traffic data completion to enhance data quality, (ii) a traffic domain adapter for uniform traffic data transformation to address data distribution discrepancies, and (iii) a traffic secret aggregation protocol for secure traffic data aggregation to safeguard data privacy. Extensive experiments on 4 real-world datasets demonstrate that the proposed FedTT framework outperforms the 14 state-of-the-art baselines.

cs.LG↗

arXiv2026

PACER: Blockwise Pre-verification for Speculative Decoding with Adaptive Length

Speculative decoding (SD) is a powerful technique for accelerating the inference process of large language models (LLMs) without sacrificing accuracy. Typically, SD employs a small draft model to generate a fixed number of draft tokens, which are then verified in parallel by the target model. However, our experiments reveal that the optimal draft length varies significantly across different decoding steps. This variation suggests that using a fixed draft length limits the potential for further improvements in decoding speed. To address this challenge, we propose Pacer, a novel approach that dynamically controls draft length using a lightweight, trainable pre-verification layer. This layer pre-verifies draft tokens blockwise before they are sent to the target model, allowing the draft model to stop token generation if the blockwise pre-verification fails. We implement Pacer on multiple SD model pairs and evaluate its performance across various benchmarks. Our results demonstrate that Pacer achieves up to 2.66x Speedup over autoregressive decoding and consistently outperforms standard speculative decoding. Furthermore, when integrated with Ouroboros, Pacer attains up to 3.09x Speedup.

cs.CL↗

arXiv2026

Shape-morphing programming of soft materials on complex geometries via neural operator

Shape-morphing soft materials can enable diverse target morphologies through voxel-level material distribution design, offering significant potential for various applications. Despite progress in basic shape-morphing design with simple geometries, achieving advanced applications such as conformal implant deployment or aerodynamic morphing requires accurate and diverse morphing designs on complex geometries, which remains challenging. Here, we present a Spectral and Spatial Neural Operator (S2NO), which enables high-fidelity morphing prediction on complex geometries. S2NO effectively captures global and local morphing behaviours on irregular computational domains by integrating Laplacian eigenfunction encoding and spatial convolutions. Combining S2NO with evolutionary algorithms enables voxel-level optimisation of material distributions for shape morphing programming on various complex geometries, including irregular-boundary shapes, porous structures, and thin-walled structures. Furthermore, the neural operator's discretisation-invariant property enables super-resolution material distribution design, further expanding the diversity and complexity of morphing design. These advancements significantly improve the efficiency and capability of programming complex shape morphing.

cs.LG↗

arXiv2026

PhyG-MoE: A Physics-Guided Mixture-of-Experts Framework for Energy-Efficient GNSS Interference Recognition

Complex electromagnetic interference increasingly compromises Global Navigation Satellite Systems (GNSS), threatening the reliability of Space-Air-Ground Integrated Networks (SAGIN). Although deep learning has advanced interference recognition, current static models suffer from a \textbf{fundamental limitation}: they impose a fixed computational topology regardless of the input's physical entropy. This rigidity leads to severe resource mismatch, where simple primitives consume the same processing cost as chaotic, saturated mixtures. To resolve this, this paper introduces PhyG-MoE (Physics-Guided Mixture-of-Experts), a framework designed to \textbf{dynamically align model capacity with signal complexity}. Unlike static architectures, the proposed system employs a spectrum-based gating mechanism that routes signals based on their spectral feature entanglement. A high-capacity TransNeXt expert is activated on-demand to disentangle complex features in saturated scenarios, while lightweight experts handle fundamental signals to minimize latency. Evaluations on 21 jamming categories demonstrate that PhyG-MoE achieves an overall accuracy of 97.58\%. By resolving the intrinsic conflict between static computing and dynamic electromagnetic environments, the proposed framework significantly reduces computational overhead without performance degradation, offering a viable solution for resource-constrained cognitive receivers.

cs.CV↗

arXiv2026

PaperGuide: Making Small Language-Model Paper-Reading Agents More Efficient

The accelerating growth of the scientific literature makes it increasingly difficult for researchers to track new advances through manual reading alone. Recent progress in large language models (LLMs) has therefore spurred interest in autonomous agents that can read scientific papers and extract task-relevant information. However, most existing approaches rely either on heavily engineered prompting or on a conventional SFT-RL training pipeline, both of which often lead to excessive and low-yield exploration. Drawing inspiration from cognitive science, we propose PaperCompass, a framework that mitigates these issues by separating high-level planning from fine-grained execution. PaperCompass first drafts an explicit plan that outlines the intended sequence of actions, and then performs detailed reasoning to instantiate each step by selecting the parameters for the corresponding function calls. To train such behavior, we introduce Draft-and-Follow Policy Optimization (DFPO), a tailored RL method that jointly optimizes both the draft plan and the final solution. DFPO can be viewed as a lightweight form of hierarchical reinforcement learning, aimed at narrowing the `knowing-doing' gap in LLMs. We provide a theoretical analysis that establishes DFPO's favorable optimization properties, supporting a stable and reliable training process. Experiments on paper-based question answering (Paper-QA) benchmarks show that PaperCompass improves efficiency over strong baselines without sacrificing performance, achieving results comparable to much larger models.

cs.LG↗

arXiv2026

JSR-GFNet: Jamming-to-Signal Ratio-Aware Dynamic Gating for Interference Classification in future Cognitive Global Navigation Satellite Systems

The transition toward cognitive global navigation satellite system (GNSS) receivers requires accurate interference classification to trigger adaptive mitigation strategies. However, conventional methods relying on Time-Frequency Analysis (TFA) and Convolutional Neural Networks (CNNs) face two fundamental limitations: severe performance degradation in low Jamming-to-Signal Ratio (JSR) regimes due to noise obscuration, and ``feature degeneracy'' caused by the loss of phase information in magnitude-only spectrograms. Consequently, spectrally similar signals -- such as high-order Quadrature Amplitude Modulation versus Band-Limited Gaussian Noise -- become indistinguishable. To overcome these challenges, this paper proposes the \textbf{JSR-Guided Fusion Network (JSR-GFNet)}. This multi-modal architecture combines phase-sensitive complex In-Phase/Quadrature (IQ) samples with Short-Time Fourier Transform (STFT) spectrograms. Central to this framework is a physics-inspired dynamic gating mechanism driven by statistical signal descriptors. Acting as a conditional controller, it autonomously estimates signal reliability to dynamically reweight the contributions of a Complex-Valued ResNet (IQ stream) and an EfficientNet backbone (STFT stream). To validate the model, we introduce the Comprehensive GNSS Interference (CGI-21) dataset, simulating 21 jamming categories including software-defined waveforms from aerial platforms. Extensive experiments demonstrate that JSR-GFNet achieves higher accuracy across the full 10--50 dB JSR spectrum. Notably, interpretability analysis confirms that the model learns a physically intuitive strategy: prioritizing spectral energy integration in noise-limited regimes while shifting focus to phase precision in high-SNR scenarios to resolve modulation ambiguities. This framework provides a robust solution for next-generation aerospace navigation security.

eess.SP↗

arXiv2026

DeXOR: Enabling XOR in Decimal Space for Streaming Lossless Compression of Floating-point Data

With streaming floating-point numbers being increasingly prevalent, effective and efficient compression of such data is critical. Compression schemes must be able to exploit the similarity, or smoothness, of consecutive numbers and must be able to contend with extreme conditions, such as high-precision values or the absence of smoothness. We present DeXOR, a novel framework that enables decimal XOR procedure to encode decimal-space longest common prefixes and suffixes, achieving optimal prefix reuse and effective redundancy elimination. To ensure accurate and low-cost decompression even with binary-decimal conversion errors, DeXOR incorporates 1) scaled truncation with error-tolerant rounding and 2) different bit management strategies optimized for decimal XOR. Additionally, a robust exception handler enhances stability by managing floating-point exponents, maintaining high compression ratios under extreme conditions. In evaluations across 22 datasets, DeXOR surpasses state-of-the-art schemes, achieving a 15% higher compression ratio and a 20% faster decompression speed while maintaining a competitive compression speed. DeXOR also offers scalability under varying conditions and exhibits robustness in extreme scenarios where other schemes fail.

cs.DB↗

arXiv2025

Exploring Efficiency Frontiers of Thinking Budget in Medical Reasoning: Scaling Laws between Computational Resources and Reasoning Quality

This study presents the first comprehensive evaluation of thinking budget mechanisms in medical reasoning tasks, revealing fundamental scaling laws between computational resources and reasoning quality. We systematically evaluated two major model families, Qwen3 (1.7B to 235B parameters) and DeepSeek-R1 (1.5B to 70B parameters), across 15 medical datasets spanning diverse specialties and difficulty levels. Through controlled experiments with thinking budgets ranging from zero to unlimited tokens, we establish logarithmic scaling relationships where accuracy improvements follow a predictable pattern with both thinking budget and model size. Our findings identify three distinct efficiency regimes: high-efficiency (0 to 256 tokens) suitable for real-time applications, balanced (256 to 512 tokens) offering optimal cost-performance tradeoffs for routine clinical support, and high-accuracy (above 512 tokens) justified only for critical diagnostic tasks. Notably, smaller models demonstrate disproportionately larger benefits from extended thinking, with 15 to 20% improvements compared to 5 to 10% for larger models, suggesting a complementary relationship where thinking budget provides greater relative benefits for capacity-constrained models. Domain-specific patterns emerge clearly, with neurology and gastroenterology requiring significantly deeper reasoning processes than cardiovascular or respiratory medicine. The consistency between Qwen3 native thinking budget API and our proposed truncation method for DeepSeek-R1 validates the generalizability of thinking budget concepts across architectures. These results establish thinking budget control as a critical mechanism for optimizing medical AI systems, enabling dynamic resource allocation aligned with clinical needs while maintaining the transparency essential for healthcare deployment.

cs.CL↗

arXiv2025

47B Mixture-of-Experts Beats 671B Dense Models on Chinese Medical Examinations

The rapid advancement of large language models(LLMs) has prompted significant interest in their potential applications in medical domains. This paper presents a comprehensive benchmark evaluation of 27 state-of-the-art LLMs on Chinese medical examination questions, encompassing seven medical specialties across two professional levels. We introduce a robust evaluation framework that assesses model performance on 2,800 carefully curated questions from cardiovascular, gastroenterology, hematology, infectious diseases, nephrology, neurology, and respiratory medicine domains. Our dataset distinguishes between attending physician and senior physician difficulty levels, providing nuanced insights into model capabilities across varying complexity. Our empirical analysis reveals substantial performance variations among models, with Mixtral-8x7B achieving the highest overall accuracy of 74.25%, followed by DeepSeek-R1-671B at 64.07%. Notably, we observe no consistent correlation between model size and performance, as evidenced by the strong performance of smaller mixture-of-experts architectures. The evaluation demonstrates significant performance gaps between medical specialties, with models generally performing better on cardiovascular and neurology questions compared to gastroenterology and nephrology domains. Furthermore, our analysis indicates minimal performance degradation between attending and senior physician levels for top-performing models, suggesting robust generalization capabilities. This benchmark provides critical insights for the deployment of LLMs in medical education and clinical decision support systems, highlighting both the promise and current limitations of these technologies in specialized medical contexts.

cs.CL↗