arXiv subjects

Rui Zhang

Publications and source records attributed to Rui Zhang.

At least 199 records · Page 11Linked to original sources

arXiv2026

Learning to Jointly Optimize Antenna Positioning and Beamforming for Movable Antenna-Aided Systems

The recently emerged movable antenna (MA) and fluid antenna technologies offer promising solutions to enhance the spatial degrees of freedom in wireless systems by dynamically adjusting the positions of transmit or receive antennas within given regions. In this paper, we aim to address the joint optimization problem of antenna positioning and beamforming in MA-aided multi-user downlink transmission systems. This problem involves mixed discrete antenna position and continuous beamforming weight variables, along with coupled distance constraints on antenna positions, which pose significant challenges for optimization algorithm design. To overcome these challenges, we propose an end-to-end deep learning framework, consisting of a positioning model that handles the discrete variables and the coupled constraints, and a beamforming model that handles the continuous variables. Simulation results demonstrate that the proposed framework achieves superior sum rate performance, yet with much reduced computation time compared to existing methods.

eess.SP↗

arXiv2026

Directivity Enhancement of Movable Antenna Arrays with Mutual Coupling

In conventional antenna arrays, mutual coupling between antenna elements is often regarded as detrimental. However, under specific conditions, it can be harnessed to enhance the far-field directivity (i.e., beamforming gain). Theoretically, the directivity of an N-antenna superdirective array over the endfire direction can reach N^{2}, significantly exceeding the directivity of a traditional uncoupled array which is N over all directions. This paper investigates the potential of mutual coupling effects in movable antenna (MA) arrays for directivity enhancement. A low-complexity algorithm called Greedy Search and Gradient Descent (GS-GD) is proposed to optimize the antenna positions for maximizing the array directivity over any given direction, where the antenna positions are first selected sequentially from discrete grid points and then continuously refined through gradient descent (GD) optimization. Numerical results demonstrate that the optimized MA array design by exploiting the antenna coupling achieves significant directivity gains compared to the conventional uniform linear array (ULA) without antenna coupling over all directions. Additionally, the proposed GS-GD algorithm is shown to approach the global optimum closely in most directions.

cs.IT↗

arXiv2026

MedCL-Bench: Benchmarking stability-efficiency trade-offs and scaling in biomedical continual learning

Medical language models must be updated as evidence and terminology evolve, yet sequential updating can trigger catastrophic forgetting. Although biomedical NLP has many static benchmarks, no unified, task-diverse benchmark exists for evaluating continual learning under standardized protocols, robustness to task order and compute-aware reporting. We introduce MedCL-Bench, which streams ten biomedical NLP datasets spanning five task families and evaluates eleven continual learning strategies across eight task orders, reporting retention, transfer, and GPU-hour cost. Across backbones and task orders, direct sequential fine-tuning on incoming tasks induces catastrophic forgetting, causing update-induced performance regressions on prior tasks. Continual learning methods occupy distinct retention-compute frontiers: parameter-isolation provides the best retention per GPU-hour, replay offers strong protection at higher cost, and regularization yields limited benefit. Forgetting is task-dependent, with multi-label topic classification most vulnerable and constrained-output tasks more robust. MedCL-Bench provides a reproducible framework for auditing model updates before deployment.

cs.AI↗

arXiv2026

Hydrodynamic Modeling of Odd Nematic Elasticity in Liquid Crystals

There is a recent interest in studying odd elasticity in soft solids. Current focus has been on simple solids. However, many soft solids are structured and can exhibit nematic elasticity or viscoelasticity. Here we generalize the concept of odd elasticity to nematic elasticity. By rewriting the governing equation for two-dimensional nematic liquid crystals (LCs) in terms of complex Ginzburg--Landau equation, we propose an odd nematic elastic term and its stress term in the hydrodynamic model of nematic LCs. The odd nematic elasticity can be physically interpreted as non-reciprocal interactions between neighboring directors. In odd nematics we find that domain walls become self-propelled and are accompanied by a bidirectional flow, and point defects can self-spin, develop a spiral pattern, and induce a vortical flow. Interactions of a pair of defects show rich dynamics that are distinct from those in active nematics. As such, we have developed an odd general elasticity, which can be further generalized to other viscoelastic materials, and proposed a novel way to manipulate topological defects in nematic LCs.

cond-mat.soft↗

arXiv2026

QiMeng-CodeV-SVA: Training Specialized LLMs for Hardware Assertion Generation via RTL-Grounded Bidirectional Data Synthesis

SystemVerilog Assertions (SVAs) are crucial for hardware verification. Recent studies leverage general-purpose LLMs to translate natural language properties to SVAs (NL2SVA), but they perform poorly due to limited data. We propose a data synthesis framework to tackle two challenges: the scarcity of high-quality real-world SVA corpora and the lack of reliable methods to determine NL-SVA semantic equivalence. For the former, large-scale open-source RTLs are used to guide LLMs to generate real-world SVAs; for the latter, bidirectional translation serves as a data selection method. With the synthesized data, we train CodeV-SVA, a series of SVA generation models. Notably, CodeV-SVA-14B achieves 75.8% on NL2SVA-Human and 84.0% on NL2SVA-Machine in Func.@1, matching or exceeding advanced LLMs like GPT-5 and DeepSeek-R1.

cs.CL↗

arXiv2026

3-D Trajectory Optimization for Robust Direction Sensing in Movable Antenna Systems

This paper presents a novel wireless sensing system where a movable antenna (MA) continuously moves and receives sensing signals within a three-dimensional (3-D) region to enhance sensing performance compared with conventional fixed-position antenna (FPA)-based sensing. We show that the performance of direction vector estimation for a target is fundamentally related to the 3-D MA trajectory in terms of the mean square angular error lower-bound (MSAEB), which is adopted as a coordinate-invariant performance metric. In particular, the closed-form expression of the MSAEB is derived as a function of the trajectory covariance matrix. Theoretical analysis shows that two-dimensional (2-D) antenna movement suffers from performance divergence for target direction close to the endfire direction of the 2-D MA plane, whereas 3-D movement can achieve isotropic sensing performance over the entire angular region. To achieve robust sensing performance, we formulate a min-max optimization problem to minimize the maximum (worst-case) MSAEB over a given continuous angular region wherein the target is located. An efficient successive convex approximation (SCA) algorithm is developed to optimize the 3-D MA trajectory and obtain a locally optimal solution. Numerical results demonstrate that the proposed 3-D MA sensing scheme is able to significantly reduce the worst-case mean square angular error (MSAE) compared with conventional arrays with FPAs and MA systems with 2-D movement only, thus achieving more accurate and robust direction estimation over the entire angular region.

cs.IT↗

arXiv2026

HeartAgent: An Autonomous Agent System for Explainable Differential Diagnosis in Cardiology

Heart diseases remain a leading cause of morbidity and mortality worldwide, necessitating accurate and trustworthy differential diagnosis. However, existing artificial intelligence-based diagnostic methods are often limited by insufficient cardiology knowledge, inadequate support for complex reasoning, and poor interpretability. Here we present HeartAgent, a cardiology-specific agent system designed to support a reliable and explainable differential diagnosis. HeartAgent integrates customized tools and curated data resources and orchestrates multiple specialized sub-agents to perform complex reasoning while generating transparent reasoning trajectories and verifiable supporting references. Evaluated on the MIMIC dataset and a private electronic health records cohort, HeartAgent achieved over 36% and 20% improvements over established comparative methods, in top-3 diagnostic accuracy, respectively. Additionally, clinicians assisted by HeartAgent demonstrated gains of 26.9% in diagnostic accuracy and 22.7% in explanatory quality compared with unaided experts. These results demonstrate that HeartAgent provides reliable, explainable, and clinically actionable decision support for cardiovascular care.

cs.CL↗

arXiv2026

Compose Your Policies! Improving Diffusion-based or Flow-based Robot Policies via Test-time Distribution-level Composition

Diffusion-based models for robotic control, including vision-language-action (VLA) and vision-action (VA) policies, have demonstrated significant capabilities. Yet their advancement is constrained by the high cost of acquiring large-scale interaction datasets. This work introduces an alternative paradigm for enhancing policy performance without additional model training. Perhaps surprisingly, we demonstrate that the composed policies can exceed the performance of either parent policy. Our contribution is threefold. First, we establish a theoretical foundation showing that the convex composition of distributional scores from multiple diffusion models can yield a superior one-step functional objective compared to any individual score. A Grönwall-type bound is then used to show that this single-step improvement propagates through entire generation trajectories, leading to systemic performance gains. Second, motivated by these results, we propose General Policy Composition (GPC), a training-free method that enhances performance by combining the distributional scores of multiple pre-trained policies via a convex combination and test-time search. GPC is versatile, allowing for the plug-and-play composition of heterogeneous policies, including VA and VLA models, as well as those based on diffusion or flow-matching, irrespective of their input visual modalities. Third, we provide extensive empirical validation. Experiments on Robomimic, PushT, and RoboTwin benchmarks, alongside real-world robotic evaluations, confirm that GPC consistently improves performance and adaptability across a diverse set of tasks. Further analysis of alternative composition operators and weighting strategies offers insights into the mechanisms underlying the success of GPC. These results establish GPC as a simple yet effective method for improving control performance by leveraging existing policies.

cs.RO↗

arXiv2026

Estimation of heterogeneous principal effects under principal ignorability

We study estimation and inference for heterogeneous principal causal effects with binary treatments and binary intermediate variables. Principal causal effects are subgroup effects within strata defined by potential values of an intermediate variable, including effects among compliers. We propose a framework for estimating and forming pointwise confidence intervals for heterogeneous principal causal effects under the principal ignorability assumption. Several estimators are developed, and their robustness properties are characterized: one estimator is doubly robust, whereas the other two attain intermediate robustness between double and triple robustness; in contrast, principal causal effects can be estimated in a triply robust manner only. We establish large-sample theory under nonparametric smoothness conditions and analyze the bias contributions of each approach, providing insight into performance beyond the smooth setting, including in high-dimensional regimes. Camden Coalition hotspotting randomized trial are used to illustrate the methods by estimating heterogeneous complier effects.

stat.ME↗

arXiv2026

Atomic-resolution imaging of gold species at organic liquid-solid interfaces

Understanding solid-liquid interfaces at the atomic-scale is key to improved performance of heterogeneous catalysts, electrodes and membranes. Here we combine unique specimen design, record atomic resolution in situ electron microscopy, and artificial intelligence-enabled analysis to achieve a step change in quantitative understanding of interfacial atomic behaviour. We create the first graphene liquid cells with organic solvents and employ them to track over 106 gold adatoms and clusters at a graphene surface immersed in acetone and cyclohexanone. We reveal dynamic correlated behaviour of gold adatom monomers, dimers, trimers and clusters, strongly influenced by each other, the solvent properties, and the atomic lattice of the substrate, in good agreement with theoretical calculations. We use the results to interpret differences in catalytic activity towards the industrially important acetylene hydrochlorination reaction. This new capability for exploration of atomic scale chemistry could enable rational design of future catalysts, membranes and electrodes with improved functionality.

cond-mat.mtrl-sci↗

arXiv2026

GP Bandit-Assisted Two-Stage Sparse Phase Retrieval for Amplitude-Only Near-Field Beam Training

The transition to Extremely Large Antenna Arrays (ELAA) in 6G introduces significant near-field effects, necessitating robust near-field beam training strategies in multi-path environments. Because signal phases are frequently compromised by hardware impairments such as phase noise and frequency offsets, amplitude-only channel recovery is a critical alternative to coherent beam training. However, existing near-field amplitude-based training methods often assume simplistic line-of-sight conditions. Conversely, far-field phase retrieval (PR) methods lack the sensing flexibility required to optimize training efficiency and are fundamentally limited by plane-wave models, making them ill-suited for near-field propagation. We propose a two-stage sparse PR framework for amplitude-only near-field beam training in multipath channels. Stage I performs adaptive support discovery on the standard 2D DFT beamspace by exploiting a physics-guided prior induced by near-field beam patterns. Stage II then refines the channel estimate by restricting sensing and sparse PR to the learned subspace. Numerical results show that the proposed adaptive pipeline consistently outperforms non-adaptive baselines, improving beamforming gain by over 70% at low SNR.

eess.SP↗

arXiv2026

Janus skyrmion: Interfacial quasiparticle with two-faced helicity

Janus particles are functional particles with at least two surfaces showing asymmetric properties. We show at the interface between two magnetic regions with different antisymmetric exchange interactions, an alternative species of two-dimensional topological quasiparticles can emerge, in which different helicity structures can coexist. We name such an interfacial quasiparticle a "Janus skyrmion," in analogy to the Janus particle. As the Janus skyrmion shows helicity asymmetry, its size could vary with both the in-plane and out-of-plane magnetic fields. A vertical spin current could drive the Janus skyrmion into one-dimensional motion along the interface without showing the skyrmion Hall effect, at a speed which depends on both the in-plane spin-polarization direction and current density. Thermal fluctuations could also lead to one-dimensional random walk of a Brownian Janus skyrmion. This work uncovers unique dynamics intrinsic to interfacial quasiparticles with exotic helicity, which may be realized in interface-engineered magnetic layers.

cond-mat.mes-hall↗

arXiv2026

Rotatable IRS Aided Wireless Communication

Rotatable intelligent reflecting surface (IRS) introduces a new spatial degree of freedom (DoF) by dynamically adjusting orientations without the need of changing its elements' positions in real time. To unleash the full potential of rotatable IRSs for wireless communications, this paper investigates the joint optimization of IRS rotation angles to maximize the minimum expected signal-to-noise ratio (SNR) over all locations within a given target area. We first propose an angle-dependent channel model that accurately characterizes the reception and reflection of each IRS element. Different from the conventional cosine-law assumption, the proposed model captures the practical electromagnetic characteristics of the IRS, including the effective reception area and reflection efficiency. For the single target location case, a particle swarm optimization (PSO)-based algorithm is developed to solve the SNR maximization problem, and a closed-form expression for a near-optimal solution is derived to provide useful insights. For the general area coverage enhancement case, the optimal rotation is obtained through a two-loop PSO-based iterative algorithm with null-point detection. In this algorithm, the outer loop updates the global rotation angles to maximize the minimum SNR over the target area, whereas the inner loop evaluates the SNR distribution within the area to identify the location corresponding to the minimum SNR through null-point detection. Numerical results demonstrate significant SNR improvement achieved by the proposed rotatable IRS design over various benchmark schemes under different system setups.

eess.SP↗

arXiv2026

QiMeng-CRUX: Narrowing the Gap Between Natural Language and Verilog via Core Refined Understanding eXpression for Circuit Design

Large language models (LLMs) have shown promising capabilities in hardware description language (HDL) generation. However, existing approaches often rely on free-form natural language descriptions that are often ambiguous, redundant, and unstructured, which poses significant challenges for downstream Verilog code generation. We treat hardware code generation as a complex transformation from an open-ended natural language space to a domain-specific, highly constrained target space. To bridge this gap, we introduce Core Refined Understanding eXpression (CRUX), a structured intermediate space that captures the essential semantics of user intent while organizing the expression for precise Verilog code generation. We further design a two-stage training framework, comprising Joint Expression Modeling and Dual-Space Optimization, to enhance the quality of both CRUX and Verilog code. Experiments across multiple Verilog generation benchmarks demonstrate that our model, CRUX-V, achieves state-of-the-art performance among general models, particularly under challenging design tasks. Furthermore, the CRUX space proves transferable and beneficial when used as input prompts for other code models, highlighting its effectiveness in narrowing the gap between free-form natural language descriptions and precise Verilog generation.

cs.LG↗

arXiv2026

Rethinking Code Similarity for Automated Algorithm Design with LLMs

The rise of Large Language Model-based Automated Algorithm Design (LLM-AAD) has transformed algorithm development by autonomously generating code implementations of expert-level algorithms. Unlike traditional expert-driven algorithm development, in the LLM-AAD paradigm, the main design principle behind an algorithm is often implicitly embedded in the generated code. Therefore, assessing algorithmic similarity directly from code, distinguishing genuine algorithmic innovation from mere syntactic variation, becomes essential. While various code similarity metrics exist, they fail to capture algorithmic similarity, as they focus on surface-level syntax or output equivalence rather than the underlying algorithmic logic. We propose BehaveSim, a novel method to measure algorithmic similarity through the lens of problem-solving behavior as a sequence of intermediate solutions produced during execution, dubbed as problem-solving trajectories (PSTrajs). By quantifying the alignment between PSTrajs using dynamic time warping (DTW), BehaveSim distinguishes algorithms with divergent logic despite syntactic or output-level similarities. We demonstrate its utility in two key applications: (i) Enhancing LLM-AAD: Integrating BehaveSim into existing LLM-AAD frameworks (e.g., FunSearch, EoH) promotes behavioral diversity, significantly improving performance on three AAD tasks. (ii) Algorithm analysis: BehaveSim clusters generated algorithms by behavior, enabling systematic analysis of problem-solving strategies--a crucial tool for the growing ecosystem of AI-generated algorithms. Data and code of this work are open-sourced at https://github.com/RayZhhh/behavesim.

cs.AI↗

arXiv2026

Large-Momentum Effective Theory's Asymptotic Extrapolation vs the Inverse Problem

Large-Momentum Effective Theory (LaMET) is a physics-guided systematic expansion to calculate light-cone parton distributions, including collinear (PDFs) and transverse-momentum-dependent ones, at any fixed momentum fraction $x$ within a range of $[x_{\rm min}, x_{\rm max}]$. It theoretically solves the ill-posed inverse problem that afflicts other theoretical approaches to collinear PDFs, such as short-distance factorizations. Recently, arXiv:2504.17706 [1] raised practical concerns about whether current or even future lattice data will have sufficient precision in the sub-asymptotic correlation region to support an error-controlled extrapolation -- and if not, whether it becomes an inverse problem where the relevant uncertainties cannot be properly quantified. While we agree that not all current lattice data have the desired precision to qualify for an asymptotic extrapolation, some calculations do, and more are expected in the future. We comment on the analysis and results in Ref. [1] and argue that a physics-based systematic extrapolation still provides the most reliable error estimates, even when the data quality is not ideal. In contrast, re-framing the long-distance asymptotic extrapolation as a data-driven-only inverse problem with ad hoc mathematical conditioning could lead to unnecessarily conservative errors.

hep-lat↗

arXiv2026

When Reasoning Meets Compression: Understanding the Effects of LLMs Compression on Large Reasoning Models

Compression methods, including quantization, distillation, and pruning, improve the computational efficiency of large reasoning models (LRMs). However, existing studies either fail to sufficiently compare all three compression methods on LRMs or lack in-depth interpretation analysis. In this paper, we investigate how the reasoning capabilities of LRMs are compromised during compression, through performance benchmarking and mechanistic interpretation. To uncover the effects of compression on reasoning performance, we benchmark quantized, distilled, and pruned DeepSeek-R1 models on four reasoning datasets (AIME 2024, FOLIO, Temporal Sequences, and MuSiQue). To precisely locate compression effects on model weights, we adapt difference of means and attribution patching techniques, focusing on the activation of every linear component in compressed LRMs, to interpret fine-grained causal relationships between weights and various reasoning capabilities. This fine-grained interpretation addresses a fundamental question of compression: which weights are the most important for reasoning? Overall, we find dynamically quantized 2.51-bit R1 reaches close-to-R1 performance. With empirical verification, we present three main findings that generalize across both R1 and non-R1 LRMs: (1) Weight count has a greater impact on LRMs' knowledge memorization than reasoning, highlighting the risks of pruning and distillation; (2) The MLP up projection in the final layer of distilled LRMs is one of the most important components, offering a new perspective on locating critical weights - a fundamental problem in model compression; and (3) Current quantization methods overly compress the final-layer modules and MLP gate projections, so protecting just 2% of all weights that are excessively compressed can raise average accuracy by 6.57%, greatly surpassing the state-of-the-art.

cs.LG↗

arXiv2026

FollowSpot: Enhancing Wireless Communications via Movable Ceiling-Mounted Metasurfaces

This paper studies the optimal placement of ceiling-mounted metasurfaces (MTSs) to help focus the wireless signal beam onto the target receiver, as inspired by the theatre spotlight. We assume that a total of $M$ MTSs are deployed, and that there are $L$ possible positions for each MTS. The resulting signal-to-noise (SNR) maximization problem is difficult to tackle directly because of the coupling between the placement decisions of the different MTSs. Mathematically, we are faced with a nonlinear discrete optimization problem with $L^M$ possible solutions. A remarkable result shown in this paper is that the above challenging problem can be efficiently solved within $O(ML^2\log(ML))$ time. There are two key steps in developing the proposed algorithm. First, we successfully decouple the placement variables of different MTSs by introducing a continuous auxiliary variable $μ$; the discrete primal variables are now easy to optimize when $μ$ is held fixed, but the optimization problem of $μ$ is nonconvex. Second, we show that the optimization of continuous $μ$ can be recast into a discrete optimization problem with only $LM$ possible solutions, so the optimal $μ$ can now be readily obtained. Numerical results show that the proposed algorithm can not only guarantee a global optimum but also reach the optimal solution efficiently.

cs.IT↗