arXiv subjects

Ruiqi Zhang

Publications and source records attributed to Ruiqi Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

SPEED-RL: Faster Training of Reasoning Models via Online Curriculum Learning

Training large language models with reinforcement learning (RL) against verifiable rewards significantly enhances their reasoning abilities, yet remains computationally expensive due to inefficient uniform prompt sampling. We introduce Selective Prompting with Efficient Estimation of Difficulty (SPEED), an adaptive online RL curriculum that selectively chooses training examples of intermediate difficulty to maximize learning efficiency. Theoretically, we establish that intermediate-difficulty prompts improve the gradient estimator's signal-to-noise ratio, accelerating convergence. Empirically, our efficient implementation leads to 2x to 6x faster training without degrading accuracy, requires no manual tuning, and integrates seamlessly into standard RL algorithms.

cs.LG↗

arXiv2026

Rotational Phonons Drive Low-Energy Kinks in Cuprate Superconductors

Angle-resolved photoemission spectroscopy (ARPES) reveals ubiquitous quasiparticle ``kinks'' near $\sim$70 meV and $\sim$40 meV across cuprate superconductors, often accompanied by peak--dip--hump (PDH) structures. These features point to strong coupling between electrons and low-energy bosonic excitations, but the microscopic origin has remained elusive due to the limitations of conventional density-functional theory (DFT) and the high cost of beyond-DFT methods. Here, we systematically study the electron--phonon coupling (EPC) in hole-doped infinite-layer CaCuO$_2$ using the Strongly Constrained and Appropriately Normed (SCAN) density functional, explicitly including magnetic effects. We find a substantial EPC strength $λ$ of $\sim$0.5 in the magnetic phase, producing kinks and PDH structures in the 40-80~meV window in excellent agreement with experiments. The dominant contribution arises from rotational oxygen phonons, while breathing modes contribute little. Our results establish strong EPC in cuprates, highlight the key role of rotational phonons, and provide a framework for understanding spectral anomalies in cuprates and beyond.

cond-mat.str-el↗

arXiv2026

Mesh Splatting for End-to-end Multiview Surface Reconstruction

Surfaces are typically represented as meshes, which can be extracted from volumetric fields via meshing or optimized directly as surface parameterizations. Volumetric representations occupy 3D space and have a large effective receptive field along rays, enabling stable and efficient optimization via volumetric rendering; however, subsequent meshing often produces overly dense meshes and introduces accumulated errors. In contrast, pure surface methods avoid meshing but capture only boundary geometry with a single-layer receptive field, making it difficult to learn intricate geometric details and increasing reliance on priors (e.g., shading or normals). We bridge this gap by differentiably turning a surface representation into a volumetric one, enabling end-to-end surface reconstruction via volumetric rendering to model complex geometries. Specifically, we soften a mesh into multiple semi-transparent layers that remain differentiable with respect to the base mesh, endowing it with a controllable 3D receptive field. Combined with a splatting-based renderer and a topology-control strategy, our method can be optimized in about 20 minutes to achieve accurate surface reconstruction while substantially improving mesh quality.

cs.GR↗

arXiv2026

Enhancing classical simulation with noisy quantum devices

As quantum devices continue to improve in scale and precision, a central challenge is how to effectively utilize noisy hardware for meaningful computation. Most existing approaches aim to recover noiseless circuit outputs from noisy ones through error mitigation or correction. Here, we show that noisy quantum devices can be directly leveraged as computational resources to enhance the classical simulation of quantum circuits. We introduce the Noisy-device-enhanced Classical Simulation (NDE-CS) protocol, which improves stabilizer-based classical Monte Carlo simulation methods by incorporating data obtained from noisy quantum hardware. Specifically, NDE-CS uses noisy executions of a target circuit together with noisy Clifford circuits to learn how the target circuit can be expressed in terms of Clifford circuits under realistic noise. The same learned relation can then be reused in the noiseless Clifford limit, enabling accurate estimation of ideal expectation values with substantially reduced sampling cost. Numerical simulations on Trotterized Ising circuits demonstrate that NDE-CS achieves orders-of-magnitude reductions in sampling cost compared to the underlying purely classical Monte Carlo approaches from which it is derived, while maintaining the same accuracy. We also compare NDE-CS with Sparse Pauli Dynamics (SPD), a powerful classical framework capable of simulating quantum circuits at previously inaccessible scales, and provide an example where the cost of SPD scales exponentially with system size, while NDE-CS scales much more favorably. These results establish NDE-CS as a scalable hybrid simulation approach for quantum circuits, where noise can be harnessed as a computational asset.

quant-ph↗

arXiv2026

A Unified Frequency Principle for Quantum and Classical Machine Learning

Quantum neural networks constitute a key class of near-term quantum learning models, yet their training dynamics remain not fully understood. Here, we present a unified theoretical framework for the frequency principle (F-principle) that characterizes the training dynamics of both classical and quantum neural networks. Within this framework, we prove that quantum neural networks exhibit a spectral bias toward learning low-frequency components of target functions, mirroring the behavior observed in classical deep networks. We further analyze the impact of noise and show that, when single-qubit noise is applied after encoding-layer rotations and modeled as a Pauli channel aligned with the rotation axis, the Fourier component labeled by $\boldsymbolω$ is suppressed by a factor $(1-2γ)^{\|\boldsymbolω\|_1}$. This leads to exponential attenuation of high-frequency terms while preserving the learnability of low-frequency structure. In the same setting, we establish that the resulting noisy circuits admit efficient classical simulation up to average-case error. Numerical experiments corroborate our theoretical predictions: Quantum neural networks primarily learn low-frequency features during early optimization and maintain robustness against dephasing and depolarizing noise acting on the encoding layer. Our results provide a frequency-domain lens that unifies classical and quantum learning dynamics, clarifies the role of noise in shaping trainability, and guides the design of noise-resilient quantum neural networks.

quant-ph↗

arXiv2025

Scalable Quantum Error Mitigation with Neighbor-Informed Learning

Noise in quantum hardware is the primary obstacle to realizing the transformative potential of quantum computing. Quantum error mitigation (QEM) offers a promising pathway to enhance computational accuracy on near-term devices, yet existing methods face a difficult trade-off between performance, resource overhead, and theoretical guarantees. In this work, we introduce neighbor-informed learning (NIL), a versatile and scalable QEM framework that unifies and strengthens existing methods such as zero-noise extrapolation (ZNE) and probabilistic error cancellation (PEC), while offering improved flexibility, accuracy, efficiency, and robustness. NIL learns to predict the ideal output of a target quantum circuit from the noisy outputs of its structurally related ``neighbor'' circuits. A key innovation is our 2-design training method, which generates training data for our machine learning model. In contrast to conventional learning-based QEM protocols that create training circuits by replacing non-Clifford gates with uniformly random Clifford gates, our approach achieves higher accuracy and efficiency, as demonstrated by both theoretical analysis and numerical simulation. Furthermore, we prove that the required size of the training set scales only \emph{logarithmically} with the total number of neighbor circuits, enabling NIL to be applied to problems involving large-scale quantum circuits. Our work establishes a theoretically grounded and practically efficient framework for QEM, paving a viable path toward achieving quantum advantage on noisy hardware.

quant-ph↗

arXiv2025

Deep Inverse Shading: Consistent Albedo and Surface Detail Recovery via Generative Refinement

Reconstructing human avatars using generative priors is essential for achieving versatile and realistic avatar models. Traditional approaches often rely on volumetric representations guided by generative models, but these methods require extensive volumetric rendering queries, leading to slow training. Alternatively, surface-based representations offer faster optimization through differentiable rasterization, yet they are typically limited by vertex count, restricting mesh resolution and scalability when combined with generative priors. Moreover, integrating generative priors into physically based human avatar modeling remains largely unexplored. To address these challenges, we introduce DIS (Deep Inverse Shading), a unified framework for high-fidelity, relightable avatar reconstruction that incorporates generative priors into a coherent surface representation. DIS centers on a mesh-based model that serves as the target for optimizing both surface and material details. The framework fuses multi-view 2D generative surface normal predictions, rich in detail but often inconsistent, into the central mesh using a normal conversion module. This module converts generative normal outputs into per-triangle surface offsets via differentiable rasterization, enabling the capture of fine geometric details beyond sparse vertex limitations. Additionally, DIS integrates a de-shading module to recover accurate material properties. This module refines albedo predictions by removing baked-in shading and back-propagates reconstruction errors to optimize the geometry. Through joint optimization of geometry and material appearance, DIS achieves physically consistent, high-quality reconstructions suitable for accurate relighting. Our experiments show that DIS delivers SOTA relighting quality, enhanced rendering efficiency, lower memory consumption, and detailed surface reconstruction.

cs.GR↗

arXiv2025

Feedback Enhancement of Time Series Aggregation for Power System Expansion Planning

As a consequence of the high variability of load demand and renewable generation, long-term and high-resolution inputs are required for power system expansion planning, making the problem intractable in real-world applications. Time series aggregation (TSA), which captures representative patterns, reduces temporal complexity while providing similar planning outputs. However, purely statistical clustering, even when enhanced with predefined ``extremes'', can overlook system-specific critical operating conditions, making it unreliable across real-world systems. Therefore, this paper links TSA accuracy on specific system operation and final solution quality, which becomes a practical bound with mean-based TSA approaches. It is observed that the distribution of operational errors is highly imbalanced, such that a few representatives dominate the total error. This paper proposes an adaptive clustering strategy based on feedback enhancement of TSA that iteratively identifies poor-performing representatives with high operational error and re-clusters only their associated periods. A study shows that the feedback enhancement improves the decision error and tighten the bound significantly compared with the plain mean-based clustering method, offering a diagnostic for TSA quality, while balancing the computational effort with solution accuracy.

math.OC↗

arXiv2025

Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning

This work studies the problem of large language model (LLM) unlearning, aiming to remove unwanted data influences (e.g., copyrighted or harmful content) while preserving model utility. Despite the increasing demand for unlearning, a technically-grounded optimization framework is lacking. Gradient ascent (GA)-type methods, though widely used, are suboptimal as they reverse the learning process without controlling optimization divergence (i.e., deviation from the pre-trained state), leading to risks of over-forgetting and potential model collapse. Negative preference optimization (NPO) has been proposed to address this issue and is considered one of the state-of-the-art LLM unlearning approaches. In this work, we revisit NPO and identify another critical issue: reference model bias. This bias arises from using the reference model (i.e., the model prior to unlearning) to evaluate the unlearning success, which can compromise NPO's effectiveness. Specifically, it leads to (a) uneven allocation of optimization power across forget data with varying difficulty levels and (b) ineffective gradient weight smoothing during the early stages of unlearning optimization. To overcome these challenges, we propose a simple yet effective unlearning optimization framework, called SimNPO, showing that `simplicity' in removing the reliance on a reference model (through the lens of simple preference optimization) benefits unlearning. We provide deeper insights into SimNPO's advantages through an analysis based on mixtures of Markov chains. Extensive experiments further validate SimNPO's efficacy on benchmarks like TOFU and MUSE, as well as its robustness against relearning attacks. Codes are available at https://github.com/OPTML-Group/Unlearn-Simple.

cs.CL↗

arXiv2025

Morphological and Chemical Changes in Cd-free Colloidal QD-LEDs During Operation

Heavy metal-free quantum-dot light-emitting devices (QD-LEDs) have demonstrated remarkable brightness, saturated color, and high efficiencies across a broad spectral range. However, in contrast to organic LEDs (OLEDs), QD-LED operational lifetimes remain limited, with the underlying degradation mechanisms not fully understood. In the present study, we show that InP/ZnSe/ZnS (red-emitting) and ZnTeSe/ZnSe/ZnS (blue-emitting) cadmium-free colloidal QD-LEDs undergo nanoscale morphological changes during operation. Specifically,interparticle coarsening and layer thinning are observed in the electron transport layer (ETL) consisting of ZnMgO nanoparticles (NPs), in the QD emissive layer, and in the organic hole transport layer. This is accompanied by the generation and diffusion of compositional oxygen- and hydrogen-radicals throughout the device, with oxygen accumulating at the electrode/ETL interfance. Moreover, in situ transmission electron microscopy reveals the electron beam exposure, in the presence of hydrogen radicals, accelerates ZnMgO NPs coarsening. To mitigate these degradation pathway, we show that acrylate-based resin-encapsulation treatment stabilize the ETL/QD layers by suppressing the radical formation and halting morphology changes. This approach achieves dramatic stability enhancements, exhibits an 8-fold and 5000-fold lifetime improvement on InP/ZnSe/ZnS and ZnTeSe/ZnSe/ZnS QD-LEDs, respectively. Our findings establish the causal relationships between the morphological degradation, interlayer radical dynamics, and state-of-the-art QD-LEDs instability, providing new insights into a scalable encapsulation treatment that enables efficient and long-lived Cd-free QD-LEDs.

physics.app-ph↗

arXiv2025

Oxygen vacancy formation in ZnSeTe blue quantum dot light-emitting diodes

Recent advancements have led to the development of bright and heavy metal-free blue-emitting quantum dot light-emitting diodes (QLEDs). However, consensus understanding of their distinct photophysical and electroluminescent dynamics remains elusive. This work correlates the chemical and electronic changes occurring in a QLED during operation using depth-resolved and operando techniques. The results indicate that oxygen vacancy forms in the ZnMgO layer during operation, with important implications on the charge injection and electrochemical dynamics. Taken together, the results suggest a causal relationship between oxygen vacancy formation and operational degradation of the blue-emitting ZnSeTe-based QLEDs.

physics.chem-ph↗

arXiv2025

Towards chemical accuracy for chemi- and physisorption with an efficient density functional

Understanding molecular adsorption on surfaces underpins many problems in chemistry and materials science. Accurately and efficiently describing the adsorption has been a challenging task for first-principles methods as the process can involve both short-range chemical bond formations and long-range physical interactions, e.g., the van der Waals (vdW) interaction. Density functional theory presents an appealing choice for modeling adsorption reactions, though calculations with many exchange correlation density functional approximations struggle to accurately describe both chemical and physical molecular adsorptions. Here, we propose an efficient density functional approximation that is accurate for both chemical and physical adsorption by concurrently optimizing its semilocal component and the long-range vdW correction against the prototypical adsorption CO/Pt(111) and Ar$_2$ binding energy curve. The resulting functional opens the door to accurate and efficient modeling of general molecular adsorption.

cond-mat.mtrl-sci↗

arXiv2025

MLMC-based Resource Adequacy Assessment with Active Learning Trained Surrogate Models

Multilevel Monte Carlo (MLMC) is a flexible and effective variance reduction technique for accelerating reliability assessments of complex power system. Recently, data-driven surrogate models have been proposed as lower-level models in the MLMC framework due to their high correlation and negligible execution time once trained. However, in resource adequacy assessments, pre-labeled datasets are typically unavailable. For large-scale systems, the efficiency gains from surrogate models are often offset by the substantial time required for labeling training data. Therefore, this paper introduces a speed metric that accounts for training time in evaluating MLMC efficiency. Considering the total time budget is limited, a vote-by-committee active learning approach is proposed to reduce the required labeling calls. A case study demonstrates that, within a given computational budget, active learning in combination with MLMC can result in a substantial reduction variance.

cs.LG↗

arXiv2025

A CPFSK Transceiver with Hybrid CSS-DSSS Spreading for LPWAN PHY Communication

Traditional low-power wide-area network (LPWAN) transceivers typically compromise data rates to achieve deep coverage. This paper presents a novel transceiver that achieves high receiver sensitivity and low computational complexity. At the transmitter, we replace the conventional direct sequence spread spectrum (DSSS) preamble with a chirp spread spectrum (CSS) preamble, consisting of a pair of down-chirp and up-chirp signals that are conjugate to each other, simplifying packet synchronization. For enhanced coverage, the payload incorporates continuous phase frequency shift keying (CPFSK) to maintain a constant envelope and phase continuity, in conjunction with DSSS to achieve a high spreading gain. At the receiver, we develop a double-peak detection method to improve synchronization and a non-coherent joint despreading and demodulation scheme that increases receiver sensitivity while maintaining simplicity in implementation. Furthermore, we optimize the preamble detection threshold and spreading sequences for maximum non-coherent receiver performance. The software-defined radio (SDR) prototype, developed using GNU Radio and USRP, along with operational snapshots, showcases its practical engineering applications. Extensive Monte Carlo simulations and field-test trials demonstrate that our transceiver outperforms traditional ones in terms of receiver sensitivity, while also being low in complexity and cost-effective for LPWAN requirements.

cs.IT↗

arXiv2025

Artificial Symmetry Breaking by Self-Interaction Error

Symmetry is a cornerstone of quantum mechanics and materials theory, underpinning the classification of electronic states and the emergence of complex phenomena such as magnetism and superconductivity. While symmetry breaking in density functional theory can reveal strong electron correlation, it may also arise spuriously from self-interaction error (SIE), an intrinsic flaw in many approximate exchange-correlation functionals. In this work, we present clear evidence that SIE alone can induce artificial symmetry breaking, even in the absence of strong correlation. Using a family of one-electron, multi-nuclear-center systems $ \mathrm{H}^+_{n \times \frac{+2}{n}}(R) $, we show that typical semilocal density functionals exhibit symmetry-breaking localization as system size increases, deviating from the exact, symmetry-preserving Hartree-Fock solution. We further demonstrate that this localization error contrasts with the well-known delocalization error of semilocal density functionals and design a semilocal density functional that avoids the artifact. Finally, we illustrate the real-world relevance of this effect in the \ch{Ti_{Zn}v_O} defect in ZnO, where a semilocal density functional breaks the $C_{3v}$ symmetry while a hybrid density functional preserves it. These findings highlight the need for improved functional design to prevent spurious symmetry breaking in both model and real materials.

cond-mat.mtrl-sci↗

arXiv2025

Long or short CoT? Investigating Instance-level Switch of Large Reasoning Models

With the rapid advancement of large reasoning models, long Chain-of-Thought (CoT) prompting has demonstrated strong performance on complex tasks. However, this often comes with a significant increase in token usage. In this paper, we conduct a comprehensive empirical analysis comparing long and short CoT strategies. Our findings reveal that while long CoT can lead to performance improvements, its benefits are often marginal relative to its significantly higher token consumption. Specifically, long CoT tends to outperform when ample generation budgets are available, whereas short CoT is more effective under tighter budget constraints. These insights underscore the need for a dynamic approach that selects the proper CoT strategy based on task context and resource availability. To address this, we propose SwitchCoT, an automatic framework that adaptively chooses between long and short CoT strategies to balance reasoning accuracy and computational efficiency. Moreover, SwitchCoT is designed to be budget-aware, making it broadly applicable across scenarios with varying resource constraints. Experimental results demonstrate that SwitchCoT can reduce inference costs by up to 50% while maintaining high accuracy. Notably, under limited token budgets, it achieves performance comparable to, or even exceeding, that of using either long or short CoT alone.

cs.CL↗

arXiv2025

How Do LLMs Perform Two-Hop Reasoning in Context?

``Socrates is human. All humans are mortal. Therefore, Socrates is mortal.'' This form of argument illustrates a typical pattern of two-hop reasoning. Formally, two-hop reasoning refers to the process of inferring a conclusion by making two logical steps, each connecting adjacent concepts, such that the final conclusion depends on the integration of both steps. It is one of the most fundamental components of human reasoning and plays a crucial role in both formal logic and everyday decision-making. Despite recent progress in large language models (LLMs), we surprisingly find that they can fail at solving simple two-hop reasoning problems when distractors are present. We observe on a synthetic dataset that pre-trained LLMs often resort to random guessing among all plausible conclusions. However, after few steps of fine-tuning, models achieve near-perfect accuracy and exhibit strong length generalization. To understand the underlying mechanisms, we train a 3-layer Transformer from scratch on a synthetic two-hop reasoning task and reverse-engineer its internal information flow. We observe a clear progression in the attention logits throughout training. This pictures a sharp phase transition from an initial stage of random guessing to the emergence of a structured sequential query mechanism, where the model first retrieves the preceding and the bridge concepts in the early layers and then uses them to infer the final answer. Finally, we show that these dynamics can be captured by a minimal three-parameter attention-only network.

cs.CL↗

arXiv2025

Duawlfin: A Drone with Unified Actuation for Wheeled Locomotion and Flight Operation

This paper presents Duawlfin, a drone with unified actuation for wheeled locomotion and flight operation that achieves efficient, bidirectional ground mobility. Unlike existing hybrid designs, Duawlfin eliminates the need for additional actuators or propeller-driven ground propulsion by leveraging only its standard quadrotor motors and introducing a differential drivetrain with one-way bearings. This innovation simplifies the mechanical system, significantly reduces energy usage, and prevents the disturbance caused by propellers spinning near the ground, such as dust interference with sensors. Besides, the one-way bearings minimize the power transfer from motors to propellers in the ground mode, which enables the vehicle to operate safely near humans. We provide a detailed mechanical design, present control strategies for rapid and smooth mode transitions, and validate the concept through extensive experimental testing. Flight-mode tests confirm stable aerial performance comparable to conventional quadcopters, while ground-mode experiments demonstrate efficient slope climbing (up to 30°) and agile turning maneuvers approaching 1g lateral acceleration. The seamless transitions between aerial and ground modes further underscore the practicality and effectiveness of our approach for applications like urban logistics and indoor navigation. All the materials including 3-D model files, demonstration video and other assets are open-sourced at https://sites.google.com/view/Duawlfin.

cs.RO↗