arXiv subjects

Pei Zhang

Publications and source records attributed to Pei Zhang.

At least 55 records · Page 3Linked to original sources

arXiv2025

Pixel-Resolved Long-Context Learning for Turbulence at Exascale: Resolving Small-scale Eddies Toward the Viscous Limit

Turbulence plays a crucial role in multiphysics applications, including aerodynamics, fusion, and combustion. Accurately capturing turbulence's multiscale characteristics is essential for reliable predictions of multiphysics interactions, but remains a grand challenge even for exascale supercomputers and advanced deep learning models. The extreme-resolution data required to represent turbulence, ranging from billions to trillions of grid points, pose prohibitive computational costs for models based on architectures like vision transformers. To address this challenge, we introduce a multiscale hierarchical Turbulence Transformer that reduces sequence length from billions to a few millions and a novel RingX sequence parallelism approach that enables scalable long-context learning. We perform scaling and science runs on the Frontier supercomputer. Our approach demonstrates excellent performance up to 1.1 EFLOPS on 32,768 AMD GPUs, with a scaling efficiency of 94%. To our knowledge, this is the first AI model for turbulence that can capture small-scale eddies down to the dissipative range.

physics.flu-dyn↗

arXiv2025

LLM-based Question-Answer Framework for Sensor-driven HVAC System Interaction

Question-answering (QA) interfaces powered by large language models (LLMs) present a promising direction for improving interactivity with HVAC system insights, particularly for non-expert users. However, enabling accurate, real-time, and context-aware interactions with HVAC systems introduces unique challenges, including the integration of frequently updated sensor data, domain-specific knowledge grounding, and coherent multi-stage reasoning. In this paper, we present JARVIS, a two-stage LLM-based QA framework tailored for sensor data-driven HVAC system interaction. JARVIS employs an Expert-LLM to translate high-level user queries into structured execution instructions, and an Agent that performs SQL-based data retrieval, statistical processing, and final response generation. To address HVAC-specific challenges, JARVIS integrates (1) an adaptive context injection strategy for efficient HVAC and deployment-specific information integration, (2) a parameterized SQL builder and executor to improve data access reliability, and (3) a bottom-up planning scheme to ensure consistency across multi-stage response generation. We evaluate JARVIS using real-world data collected from a commercial HVAC system and a ground truth QA dataset curated by HVAC experts to demonstrate its effectiveness in delivering accurate and interpretable responses across diverse queries. Results show that JARVIS consistently outperforms baseline and ablation variants in both automated and user-centered assessments, achieving high response quality and accuracy.

cs.AI↗

arXiv2025

Multi-task parallelism for robust pre-training of graph foundation models on multi-source, multi-fidelity atomistic modeling data

Graph foundation models using graph neural networks promise sustainable, efficient atomistic modeling. To tackle challenges of processing multi-source, multi-fidelity data during pre-training, recent studies employ multi-task learning, in which shared message passing layers initially process input atomistic structures regardless of source, then route them to multiple decoding heads that predict data-specific outputs. This approach stabilizes pre-training and enhances a model's transferability to unexplored chemical regions. Preliminary results on approximately four million structures are encouraging, yet questions remain about generalizability to larger, more diverse datasets and scalability on supercomputers. We propose a multi-task parallelism method that distributes each head across computing resources with GPU acceleration. Implemented in the open-source HydraGNN architecture, our method was trained on over 24 million structures from five datasets and tested on the Perlmutter, Aurora, and Frontier supercomputers, demonstrating efficient scaling on all three highly heterogeneous super-computing architectures.

cs.LG↗

arXiv2025

In-field Calibration of Low-Cost Sensors through XGBoost $\&$ Aggregate Sensor Data

Effective large-scale air quality monitoring necessitates distributed sensing due to the pervasive and harmful nature of particulate matter (PM), particularly in urban environments. However, precision comes at a cost: highly accurate sensors are expensive, limiting the spatial deployments and thus their coverage. As a result, low-cost sensors have become popular, though they are prone to drift caused by environmental sensitivity and manufacturing variability. This paper presents a model for in-field sensor calibration using XGBoost ensemble learning to consolidate data from neighboring sensors. This approach reduces dependence on the presumed accuracy of individual sensors and improves generalization across different locations.

cs.LG↗

arXiv2025

ConText: Driving In-context Learning for Text Removal and Segmentation

This paper presents the first study on adapting the visual in-context learning (V-ICL) paradigm to optical character recognition tasks, specifically focusing on text removal and segmentation. Most existing V-ICL generalists employ a reasoning-as-reconstruction approach: they turn to using a straightforward image-label compositor as the prompt and query input, and then masking the query label to generate the desired output. This direct prompt confines the model to a challenging single-step reasoning process. To address this, we propose a task-chaining compositor in the form of image-removal-segmentation, providing an enhanced prompt that elicits reasoning with enriched intermediates. Additionally, we introduce context-aware aggregation, integrating the chained prompt pattern into the latent query representation, thereby strengthening the model's in-context reasoning. We also consider the issue of visual heterogeneity, which complicates the selection of homogeneous demonstrations in text recognition. Accordingly, this is effectively addressed through a simple self-prompting strategy, preventing the model's in-context learnability from devolving into specialist-like, context-free inference. Collectively, these insights culminate in our ConText model, which achieves new state-of-the-art across both in- and out-of-domain benchmarks. The code is available at https://github.com/Ferenas/ConText.

cs.CV↗

arXiv2025

STORM-BORN: A Challenging Mathematical Derivations Dataset Curated via a Human-in-the-Loop Multi-Agent Framework

High-quality math datasets are crucial for advancing the reasoning abilities of large language models (LLMs). However, existing datasets often suffer from three key issues: outdated and insufficient challenging content, neglecting human-like reasoning, and limited reliability due to single-LLM generation. To address these, we introduce STORM-BORN, an ultra-challenging dataset of mathematical derivations sourced from cutting-edge academic papers, which includes dense human-like approximations and heuristic cues. To ensure the reliability and quality, we propose a novel human-in-the-loop, multi-agent data generation framework, integrating reasoning-dense filters, multi-agent collaboration, and human mathematicians' evaluations. We curated a set of 2,000 synthetic samples and deliberately selected the 100 most difficult problems. Even most advanced models like GPT-o1 solved fewer than 5% of them. Fine-tuning on STORM-BORN boosts accuracy by 7.84% (LLaMA3-8B) and 9.12% (Qwen2.5-7B). As AI approaches mathematician-level reasoning, STORM-BORN provides both a high-difficulty benchmark and a human-like reasoning training resource. Our code and dataset are publicly available at https://github.com/lwhere/STORM-BORN.

cs.CL↗

arXiv2025

Preliminary Characterization of Bio-inspired Dog-Nose Sampler for Aerosol Detection

Before aerosols can be sensed, sampling technologies must capture the particulate matter of interest. To that end, for systems deployed in open environments where the location of the aerosol is unknown, extending the reach of the sampler could lessen the precision required in sensor placement or reduce the number of sensors required for full spatial coverage. Inspired by the sensitivity of the canine olfactory system, this paper presents a rudimentary sampler that mimics the air flow of a dog's nose. The design consists of speed-controlled inhalation jets, as well as exhalation jets that are angled down and to the side. We tested this design on volatile organic compounds (VOC) in a small number of scenarios to validate the concept and understand how the system behaves. We show that in preliminary testing this dog-nose setup provides improvements over passive and solely inhalation sensing.

eess.SY↗

arXiv2025

ZV-Sim: Probabilistic Simulation Framework for Pre-emergent Novel Zoonose Tracking

ZV-Sim is an open-source, modular Python framework for probabilistic simulation and analysis of pre-emergent novel zoonotic diseases using pervasive sensing data. It incorporates customizable Human and Animal Presence agents that leverage known and simulated location data, contact networks, and illness reports to assess and predict disease origins and spread. The framework supports Monte Carlo experiments to analyze outcomes with various user-defined movement and probability models. Although initial models are basic and illustrative, ZV-Sim's extensible design facilitates the integration of more sophisticated models as richer data become available, enhancing future capabilities in zoonotic disease tracking. The source code is publicly available \href{https://github.com/jmaff/zv-sim}{\underline{\textit{here}}}.

q-bio.QM↗

arXiv2025

A Separable Self-attention Inspired by the State Space Model for Computer Vision

Mamba is an efficient State Space Model (SSM) with linear computational complexity. Although SSMs are not suitable for handling non-causal data, Vision Mamba (ViM) methods still demonstrate good performance in tasks such as image classification and object detection. Recent studies have shown that there is a rich theoretical connection between state space models and attention variants. We propose a novel separable self attention method, for the first time introducing some excellent design concepts of Mamba into separable self-attention. To ensure a fair comparison with ViMs, we introduce VMINet, a simple yet powerful prototype architecture, constructed solely by stacking our novel attention modules with the most basic down-sampling layers. Notably, VMINet differs significantly from the conventional Transformer architecture. Our experiments demonstrate that VMINet has achieved competitive results on image classification and high-resolution dense prediction tasks.Code is available at: https://github.com/yws-wxs/VMINet.

cs.CV↗

arXiv2025

Qwen3 Technical Report

In this work, we present Qwen3, the latest version of the Qwen model family. Qwen3 comprises a series of large language models (LLMs) designed to advance performance, efficiency, and multilingual capabilities. The Qwen3 series includes models of both dense and Mixture-of-Expert (MoE) architectures, with parameter scales ranging from 0.6 to 235 billion. A key innovation in Qwen3 is the integration of thinking mode (for complex, multi-step reasoning) and non-thinking mode (for rapid, context-driven responses) into a unified framework. This eliminates the need to switch between different models--such as chat-optimized models (e.g., GPT-4o) and dedicated reasoning models (e.g., QwQ-32B)--and enables dynamic mode switching based on user queries or chat templates. Meanwhile, Qwen3 introduces a thinking budget mechanism, allowing users to allocate computational resources adaptively during inference, thereby balancing latency and performance based on task complexity. Moreover, by leveraging the knowledge from the flagship models, we significantly reduce the computational resources required to build smaller-scale models, while ensuring their highly competitive performance. Empirical evaluations demonstrate that Qwen3 achieves state-of-the-art results across diverse benchmarks, including tasks in code generation, mathematical reasoning, agent tasks, etc., competitive against larger MoE models and proprietary models. Compared to its predecessor Qwen2.5, Qwen3 expands multilingual support from 29 to 119 languages and dialects, enhancing global accessibility through improved cross-lingual understanding and generation capabilities. To facilitate reproducibility and community-driven research and development, all Qwen3 models are publicly accessible under Apache 2.0.

cs.CL↗

arXiv2025

Magnetic Field induced control and Multiple Magnomechanically Induced Transparency in Single Cavity

We investigate magnomechanically induced transparency (MMIT) in a microwave 3D copper cavity with two YIG spheres under varying interaction parameters. Numerical simulations show that the steady-state magnon number increases with stronger coupling between cavity photons and magnons, and is sensitive to both bias and drive magnetic fields. Pronounced peaks in the magnon population near resonant fields highlight the importance of the bias field in energy transfer. The transparency windows are tunable, with up to quadruple windows depending on the coupling and magnon-phonon interactions, as seen in the transmission spectrum. Dispersion analysis reveals normal and anomalous regions, enabling slow and fast light propagation modulated by coupling strength. Phase and group delay variations, influenced by the drive field, further validate the tunability of transparency windows. This study demonstrates the potential of MMIT for precise control with out any additional non-linearity over light-matter interactions, with applications in quantum information processing and optical communications.

physics.optics↗

arXiv2025

Goos-Hanchen Shift and Slow Light Enhancement in a Fixed Cavity: Bose-Einstein Condensate Bogoliubov Modes as Mechanical Oscillators

In this study, we explore the dynamics of slow and fast light propagation in a system consisting of a Bose-Einstein condensate (BEC) acting as a mechanical oscillator coupled to an optical parametric amplifier (OPA) within a fixed-mirror cavity. The system's response is investigated through a comprehensive analysis of the transmission spectrum, output probe field characteristics (real and imaginary components), group delay, and Goos-Hänchen shift (GHS). Our findings reveal that variations in the effective coupling strength and the OPA gain have a profound impact on the system's behavior. Specifically, as the OPA gain increases, a Fano-like resonance emerges, enhancing the transparency window and altering the dispersion, which in turn influences the group delay. The GHS is shown to be sensitive to both the incident angle and the BEC-cavity coupling strength. These results offer valuable insights into the intricate interplay between the probe field, the mechanical oscillator, and the amplified modes of the OPA, highlighting the role of these interactions in shaping the propagation of light in such systems.

physics.optics↗

arXiv2025

Falcon: Fractional Alternating Cut with Overcoming Minima in Unsupervised Segmentation

Today's unsupervised image segmentation algorithms often segment suboptimally. Modern graph-cut based approaches rely on high-dimensional attention maps from Transformer-based foundation models, typically employing a relaxed Normalized Cut solved recursively via the Fiedler vector (the eigenvector of the second smallest eigenvalue). Consequently, they still lag behind supervised methods in both mask generation speed and segmentation accuracy. We present a regularized fractional alternating cut (Falcon), an optimization-based K-way Normalized Cut without relying on recursive eigenvector computations, achieving substantially improved speed and accuracy. Falcon operates in two stages: (1) a fast K-way Normalized Cut solved by extending into a fractional quadratic transformation, with an alternating iterative procedure and regularization to avoid local minima; and (2) refinement of the resulting masks using complementary low-level information, producing high-quality pixel-level segmentations. Experiments show that Falcon not only surpasses existing state-of-the-art methods by an average of 2.5% across six widely recognized benchmarks (reaching up to 4.3\% improvement on Cityscapes), but also reduces runtime by around 30% compared to prior graph-based approaches. These findings demonstrate that the semantic information within foundation-model attention can be effectively harnessed by a highly parallelizable graph cut framework. Consequently, Falcon can narrow the gap between unsupervised and supervised segmentation, enhancing scalability in real-world applications and paving the way for dense prediction-based vision pre-training in various downstream tasks. The code is released in https://github.com/KordingLab/Falcon.

cs.CV↗

arXiv2025

Leveraging Audio Representations for Vibration-Based Crowd Monitoring in Stadiums

Crowd monitoring in sports stadiums is important to enhance public safety and improve the audience experience. Existing approaches mainly rely on cameras and microphones, which can cause significant disturbances and often raise privacy concerns. In this paper, we sense floor vibration, which provides a less disruptive and more non-intrusive way of crowd sensing, to predict crowd behavior. However, since the vibration-based crowd monitoring approach is newly developed, one main challenge is the lack of training data due to sports stadiums being large public spaces with complex physical activities. In this paper, we present ViLA (Vibration Leverage Audio), a vibration-based method that reduces the dependency on labeled data by pre-training with unlabeled cross-modality data. ViLA is first pre-trained on audio data in an unsupervised manner and then fine-tuned with a minimal amount of in-domain vibration data. By leveraging publicly available audio datasets, ViLA learns the wave behaviors from audio and then adapts the representation to vibration, reducing the reliance on domain-specific vibration data. Our real-world experiments demonstrate that pre-training the vibration model using publicly available audio data (YouTube8M) achieved up to a 5.8x error reduction compared to the model without audio pre-training.

cs.SD↗

arXiv2025

Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluation

LLM self-evaluation relies on the LLM's own ability to estimate response correctness, which can greatly improve its deployment reliability. In this research track, we propose the Chain-of-Embedding (CoE) in the latent space to enable LLMs to perform output-free self-evaluation. CoE consists of all progressive hidden states produced during the inference time, which can be treated as the latent thinking path of LLMs. We find that when LLMs respond correctly and incorrectly, their CoE features differ, these discrepancies assist us in estimating LLM response correctness. Experiments in four diverse domains and seven LLMs fully demonstrate the effectiveness of our method. Meanwhile, its label-free design intent without any training and millisecond-level computational cost ensures real-time feedback in large-scale scenarios. More importantly, we provide interesting insights into LLM response correctness from the perspective of hidden state changes inside LLMs.

cs.CL↗

arXiv2025

Efficient multiplexed quantum memory with high dimensional orbital angular momentum states in cold atoms

Quantum memory plays a pivotal role in the construction of quantum repeaters, which are essential devices for establishing long-distance quantum communications and large-scale quantum networks. To boost information capacity and signal processing speed, the implementation of high-efficiency multiplexed quantum memories is essential for the development of multiplexed quantum repeaters. In this work, we experimentally demonstrate an efficient multiplexed quantum memory by consolidating photons carrying high-dimensional orbital angular momentum (OAM) state from 4 individual channels into an elongated cold $^{87}$Rb atomic ensemble. Benefiting from the cold atomic ensemble with high optical depth, we obtain a storage efficiency exceeding 70% for the 4-channel multiplexed beam. The storage fidelities surpass 83% when all channels are operated in a 4-dimensional Hilbert space, which is far above the classical benchmark and validates the quantum functionality of our memory system. The achieved high-efficiency OAM multiplexed quantum memory opens up an avenue for efficient quantum information processing over multiple parallel channels, promising significant advances in the field of quantum communication and networking.

quant-ph↗

arXiv2025

Real-time interpretation of neutron vibrational spectra with symmetry-equivariant Hessian matrix prediction

The vibrational behavior of molecules serves as a crucial fingerprint of their structure, chemical state, and surrounding environment. Neutron vibrational spectroscopy provides comprehensive measurements of vibrational modes without selection rule restrictions. However, analyzing and interpreting the resulting spectra remains a computationally formidable task. Here, we introduce a symmetry-aware neural network that directly predicts Hessian matrices from molecular structures, thereby enabling rapid vibrational spectral reconstruction. Unlike traditional approaches that focus on eigenvalue prediction, the Hessian matrix provides richer, more fundamental information with broader applications and superior extrapolation. This approach also paves the way for predicting other properties, such as reaction pathways. Trained on small molecules, our model achieves spectroscopic-level accuracy, allowing real-time, unambiguous peak assignment. Moreover, it maintains high accuracy for larger molecules, demonstrating strong transferability. This adaptability unlocks new capabilities, including on-the-fly spectral interpretation for future autonomous laboratories, and offers insights into molecular design for targeted chemical pathways.

physics.chem-ph↗

arXiv2025

WeVibe: Weight Change Estimation Through Audio-Induced Shelf Vibrations In Autonomous Stores

Weight change estimation is crucial in various applications, particularly for detecting pick-up and put-back actions when people interact with the shelf while shopping in autonomous stores. Moreover, accurate weight change estimation allows autonomous stores to automatically identify items being picked up or put back, ensuring precise cost estimation. However, the conventional approach of estimating weight changes requires specialized weight-sensing shelves, which are densely deployed weight scales, incurring intensive sensor consumption and high costs. Prior works explored the vibration-based weight sensing method, but they failed when the location of weight change varies. In response to these limitations, we made the following contributions: (1) We propose WeVibe, a first item weight change estimation system through active shelf vibration sensing. The main intuition of the system is that the weight placed on the shelf influences the dynamic vibration response of the shelf, thus altering the shelf vibration patterns. (2) We model a physics-informed relationship between the shelf vibration response and item weight across multiple locations on the shelf based on structural dynamics theory. This relationship is linear and allows easy training of a weight estimation model at a new location without heavy data collection. (3) We evaluate our system on a gondola shelf organized as the real-store settings. WeVibe achieved a mean absolute error down to 38.07g and a standard deviation of 31.2g with one sensor and 10% samples from three weight classes on estimating weight change from 0g to 450g, which can be leveraged for differentiating items with more than 100g differences.

eess.SP↗