arXiv subjects

Chen Zhang

Publications and source records attributed to Chen Zhang.

At least 289 records · Page 16Linked to original sources

arXiv2025

KAPPA: A Generic Patent Analysis Framework with Keyphrase-Based Portraits

Patent analysis highly relies on concise and interpretable document representations, referred to as patent portraits. Keyphrases, both present and absent, are ideal candidates for patent portraits due to their brevity, representativeness, and clarity. In this paper, we introduce KAPPA, an integrated framework designed to construct keyphrase-based patent portraits and enhance patent analysis. KAPPA operates in two phases: patent portrait construction and portrait-based analysis. To ensure effective portrait construction, we propose a semantic-calibrated keyphrase generation paradigm that integrates pre-trained language models with a prompt-based hierarchical decoding strategy to leverage the multi-level structural characteristics of patents. For portrait-based analysis, we develop a comprehensive framework that employs keyphrase-based patent portraits to enable efficient and accurate patent analysis. Extensive experiments on benchmark datasets of keyphrase generation, the proposed model achieves significant improvements compared to state-of-the-art baselines. Further experiments conducted on real-world patent applications demonstrate that our keyphrase-based portraits effectively capture domain-specific knowledge and enrich semantic representation for patent analysis tasks.

cs.CL↗

arXiv2025

Donald Trumps in the Virtual Polls: Simulating and Predicting Public Opinions in Surveys Using Large Language Models

In recent years, large language models (LLMs) have attracted attention due to their ability to generate human-like text. As surveys and opinion polls remain key tools for gauging public attitudes, there is increasing interest in assessing whether LLMs can accurately replicate human responses. This study examines the potential of LLMs, specifically ChatGPT-4o, to replicate human responses in large-scale surveys and to predict election outcomes based on demographic data. Employing data from the World Values Survey (WVS) and the American National Election Studies (ANES), we assess the LLM's performance in two key tasks: simulating human responses and forecasting U.S. election results. In simulations, the LLM was tasked with generating synthetic responses for various socio-cultural and trust-related questions, demonstrating notable alignment with human response patterns across U.S.-China samples, though with some limitations on value-sensitive topics. In prediction tasks, the LLM was used to simulate voting behavior in past U.S. elections and predict the 2024 election outcome. Our findings show that the LLM replicates cultural differences effectively, exhibits in-sample predictive validity, and provides plausible out-of-sample forecasts, suggesting potential as a cost-effective supplement for survey-based research.

econ.GN↗

arXiv2025

The maximal coarse Baum-Connes conjecture for spaces that admit an A-by-FCE coarse fibration structure

In this paper, we introduce a concept of A-by-FCE coarse fibration structure for metric spaces, which serves as a generalization of the A-by-CE structure for a sequence of group extensions proposed by Deng, Wang, and Yu. We prove that the maximal coarse Baum-Connes conjecture holds for metric spaces with bounded geometry that admit an A-by-FCE coarse fibration structure. As an application, the relative expanders constructed by Arzhantseva and Tessera, as well as the box spaces derived from an ``amenable-by-Haagerup'' group extension, admit the A-by-FCE coarse fibration structure. Consequently, the maximal coarse Baum-Connes conjecture holds for these spaces, which may not admit an FCE structure, i.e. fibred coarse embedding into Hilbert space.

math.KT↗

arXiv2025

A Survey on Multi-Turn Interaction Capabilities of Large Language Models

Multi-turn interaction in the dialogue system research refers to a system's ability to maintain context across multiple dialogue turns, enabling it to generate coherent and contextually relevant responses. Recent advancements in large language models (LLMs) have significantly expanded the scope of multi-turn interaction, moving beyond chatbots to enable more dynamic agentic interactions with users or environments. In this paper, we provide a focused review of the multi-turn capabilities of LLMs, which are critical for a wide range of downstream applications, including conversational search and recommendation, consultation services, and interactive tutoring. This survey explores four key aspects: (1) the core model capabilities that contribute to effective multi-turn interaction, (2) how multi-turn interaction is evaluated in current practice, (3) the general algorithms used to enhance multi-turn interaction, and (4) potential future directions for research in this field.

cs.CL↗

arXiv2025

Efficient Multiple Temporal Network Kernel Density Estimation

Kernel density estimation (KDE) has become a popular method for visual analysis in various fields, such as financial risk forecasting, crime clustering, and traffic monitoring. KDE can identify high-density areas from discrete datasets. However, most existing works only consider planar distance and spatial data. In this paper, we introduce a new model, called TN-KDE, that applies KDE-based techniques to road networks with temporal data. Specifically, we introduce a novel solution, Range Forest Solution (RFS), which can efficiently compute KDE values on spatiotemporal road networks. To support the insertion operation, we present a dynamic version, called Dynamic Range Forest Solution (DRFS). We also propose an optimization called Lixel Sharing (LS) to share similar KDE values between two adjacent lixels. Furthermore, our solutions support many non-polynomial kernel functions and still report exact values. Experimental results show that our solutions achieve up to 6 times faster than the state-of-the-art method.

cs.DB↗

arXiv2025

Data and System Perspectives of Sustainable Artificial Intelligence

Sustainable AI is a subfield of AI for concerning developing and using AI systems in ways of aiming to reduce environmental impact and achieve sustainability. Sustainable AI is increasingly important given that training of and inference with AI models such as large langrage models are consuming a large amount of computing power. In this article, we discuss current issues, opportunities and example solutions for addressing these issues, and future challenges to tackle, from the data and system perspectives, related to data acquisition, data processing, and AI model training and inference.

cs.AI↗

arXiv2025

Science objectives of the Einstein Probe mission

The Einstein Probe (EP) is an interdisciplinary mission of time-domain and X-ray astronomy. Equipped with a wide-field lobster-eye X-ray focusing imager, EP will discover cosmic X-ray transients and monitor the X-ray variability of known sources in 0.5-4 keV, at a combination of detecting sensitivity and cadence that is not accessible to the previous and current wide-field monitoring missions. EP can perform quick characterisation of transients or outbursts with a Wolter-I X-ray telescope onboard. In this paper, the science objectives of the Einstein Probe mission are presented. EP is expected to enlarge the sample of previously known or predicted but rare types of transients with a wide range of timescales. Among them, fast extragalactic transients will be surveyed systematically in soft X-rays, which include γ-ray bursts and their variants, supernova shock breakouts, and the predicted X-ray transients associated with binary neutron star mergers. EP will detect X-ray tidal disruption events and outbursts from active galactic nuclei, possibly at an early phase of the flares for some. EP will monitor the variability and outbursts of X-rays from white dwarfs, neutron stars and black holes in our and neighbouring galaxies at flux levels fainter than those detectable by the current instruments, and is expected to discover new objects. A large sample of stellar X-ray flares will also be detected and characterised. In the era of multi-messenger astronomy, EP has the potential of detecting the possible X-ray counterparts of gravitational wave events, neutrino sources, and ultra-high energy γ-ray and cosmic ray sources. EP is expected to help advance the studies of extreme objects/phenomena and their underlying physical processes revealed in the dynamic X-ray universe, as well as studies in other areas of X-ray astronomy.

astro-ph.HE↗

arXiv2025

Blueprint for Diamond Magnetometry: Unraveling Quantum Dephasing of Nitrogen-Vacancy Center Ensembles in Diamond

Diamonds with nitrogen-vacancy (NV) center ensembles are one of the most promising solid-state quantum platforms for various sensing applications. The combination of a long spin dephasing time ($T_2^*$) and a high NV center concentration is crucial for pushing the sensitivity limits. In this work, we propose a systematic measurement approach to quantify the electron spin dephasing in NV center ensembles and analyze the contributions of various sources to the dephasing time, including NV-NV interactions, strain and electric field distributions, $^{13}$C nuclear spins, and P1 electron spins. Our method is validated using a series of high-performance diamond samples, providing a comprehensive understanding of dephasing mechanisms and revealing correlations between NV concentration and different dephasing sources. Based on these insights, we further evaluate and propose strategies to improve the achievable sensitivity limits for DC magnetic field measurements.

quant-ph↗

arXiv2024

Ehrhart Polynomials of Order Polytopes: Interpreting Combinatorial Sequences on the OEIS

In this paper, we provide an overview of Ehrhart polynomials associated with order polytopes of finite posets, a concept first introduced by Stanley. We focus on their combinatorial interpretations for many sequences listed on the OEIS. We begin by exploring the Ehrhart series of order polytopes resulting from various poset operations, specifically the ordinal sum and direct sum. We then concentrate on the poset $P_λ$ associated with the Ferrers diagram of a partition $λ= (λ_1, λ_2, \ldots, λ_t)$. When $λ= (k, k-1, \ldots, 1)$, the Ehrhart polynomial is a shifted Hankel determinant of the well-known Catalan numbers; when $λ= (k, k, \ldots, k)$, the Ehrhart polynomial is solved by Stanley's hook content formula and is used to prove conjectures for the sequence [A140934] on the OEIS. When solving these problems, we rediscover Kreweras' determinant formula for the Ehrhart polynomial $\mathrm{ehr}(\mathcal{O}(P_λ), n)$ through the application of the Lindström-Gessel-Viennot lemma on non-intersecting lattice paths.

math.CO↗

arXiv2024

Long-term stability of scientific X-ray CMOS detectors

In recent years, complementary metal-oxide-semiconductor (CMOS) sensors have been demonstrated to have significant potential in X-ray astronomy, where long-term reliability is crucial for space X-ray telescopes. This study examines the long-term stability of a scientific CMOS sensor, focusing on its bias, dark current, readout noise, and X-ray spectral performance. The sensor was initially tested at -30 $^\circ$C for 16 months, followed by accelerated aging at 20 $^\circ$C. After a total aging period of 610 days, the bias map, dark current, readout noise, gain, and energy resolution exhibited no observable degradation. There are less than 50 pixels within the 4 k $\times$ 4 k array which show a decrease of the bias under 50 ms integration time by over 10 digital numbers (DNs). First-order kinetic fitting of the gain evolution predicts a gain degeneration of 0.73% over 3 years and 2.41% over 10 years. These results underscore the long-term reliability of CMOS sensors for application in space missions.

astro-ph.IM↗

arXiv2024

Understanding the RoPE Extensions of Long-Context LLMs: An Attention Perspective

Enabling LLMs to handle lengthy context is currently a research hotspot. Most LLMs are built upon rotary position embedding (RoPE), a popular position encoding method. Therefore, a prominent path is to extrapolate the RoPE trained on comparably short texts to far longer texts. A heavy bunch of efforts have been dedicated to boosting the extrapolation via extending the formulations of the RoPE, however, few of them have attempted to showcase their inner workings comprehensively. In this paper, we are driven to offer a straightforward yet in-depth understanding of RoPE extensions from an attention perspective and on two benchmarking tasks. A broad array of experiments reveals several valuable findings: 1) Maintaining attention patterns to those at the pretrained length improves extrapolation; 2) Large attention uncertainty leads to retrieval errors; 3) Using longer continual pretraining lengths for RoPE extensions could reduce attention uncertainty and significantly enhance extrapolation.

cs.CL↗

arXiv2024

ZigZagkv: Dynamic KV Cache Compression for Long-context Modeling based on Layer Uncertainty

Large Language models (LLMs) have become a research hotspot. To accelerate the inference of LLMs, storing computed caches in memory has become the standard technique. However, as the inference length increases, growing KV caches might lead to out-of-memory issues. Many existing methods address this issue through KV cache compression, primarily by preserving key tokens throughout all layers to reduce information loss. Most of them allocate a uniform budget size for each layer to retain. However, we observe that the minimum budget sizes needed to retain essential information vary across layers and models based on the perspectives of attention and hidden state output. Building on this observation, this paper proposes a simple yet effective KV cache compression method that leverages layer uncertainty to allocate budget size for each layer. Experimental results show that the proposed method can reduce memory usage of the KV caches to only $\sim$20\% when compared to Full KV inference while achieving nearly lossless performance.

cs.CL↗

arXiv2024

VoiceBench: Benchmarking LLM-Based Voice Assistants

Building on the success of large language models (LLMs), recent advancements such as GPT-4o have enabled real-time speech interactions through LLM-based voice assistants, offering a significantly improved user experience compared to traditional text-based interactions. However, the absence of benchmarks designed to evaluate these speech interaction capabilities has hindered progress of LLM-based voice assistants development. Current evaluations focus primarily on automatic speech recognition (ASR) or general knowledge evaluation with clean speeches, neglecting the more intricate, real-world scenarios that involve diverse speaker characteristics, environmental and content factors. To address this, we introduce VoiceBench, the first benchmark designed to provide a multi-faceted evaluation of LLM-based voice assistants. VoiceBench also includes both real and synthetic spoken instructions that incorporate the above three key real-world variations. Extensive experiments reveal the limitations of current LLM-based voice assistant models and offer valuable insights for future research and development in this field.

cs.CL↗

arXiv2024

COMPILED: Deep Metric Learning for Defect Classification of Threaded Pipe Connections using Multichannel Partially Observed Functional Data

In modern manufacturing, most products are conforming. Few products are nonconforming with different defect types. The identification of defect types can help further root cause diagnosis of production lines. With the sensing technology development, process variables evolved as time changes, which can be collected in high resolution as multichannel functional data. These functional data have rich information to characterize the process and help identify the defect types. Motivated by a real example from the threaded pipe connection process, we focus on defect classification where each sample is represented as partially observed multichannel functional data. However, the available samples for each defect type are limited and imbalanced. The functional data is partially observed since the pre-connection process before the threaded pipe connection process is unobserved as there is no sensor installed in the production line. Therefore, the defect classification based on imbalanced, multichannel, and partially observed functional data is very important but challenging. To deal with these challenges, we propose an innovative classification approach named as COMPILED based on deep metric learning. The framework leverages the power of deep metric learning to train on imbalanced datasets. A novel neural network structure is proposed to handle multichannel partially observed functional data. The results from a real-world case study demonstrate the superior accuracy of our framework when compared to existing benchmarks.

cs.LG↗

arXiv2024

Point-aware Interaction and CNN-induced Refinement Network for RGB-D Salient Object Detection

By integrating complementary information from RGB image and depth map, the ability of salient object detection (SOD) for complex and challenging scenes can be improved. In recent years, the important role of Convolutional Neural Networks (CNNs) in feature extraction and cross-modality interaction has been fully explored, but it is still insufficient in modeling global long-range dependencies of self-modality and cross-modality. To this end, we introduce CNNs-assisted Transformer architecture and propose a novel RGB-D SOD network with Point-aware Interaction and CNN-induced Refinement (PICR-Net). On the one hand, considering the prior correlation between RGB modality and depth modality, an attention-triggered cross-modality point-aware interaction (CmPI) module is designed to explore the feature interaction of different modalities with positional constraints. On the other hand, in order to alleviate the block effect and detail destruction problems brought by the Transformer naturally, we design a CNN-induced refinement (CNNR) unit for content refinement and supplementation. Extensive experiments on five RGB-D SOD datasets show that the proposed network achieves competitive results in both quantitative and qualitative comparisons.

cs.CV↗

arXiv2024

A Simulation Framework for Ride-Hailing with Electric Vehicles

This research presents a Python-based simulation framework designed to model electric vehicle (EV) on-demand transportation systems, with a focus on optimizing urban fleet operations. Built on a process-driven architecture, the system efficiently simulates EV fleet dynamics, including passenger matching, vehicle dispatching, and charging strategies, while enabling customization to address critical challenges such as charger placement, fleet management, and algorithm performance. We overcome the challenge of high dimensional state-space and non-Markovian system dynamics by executing processes asynchronously using SimPy and updating only the states that are affected by employing object-oriented programming. As a result, our simulation framework is capable of handling peak demand scenarios involving thousands of trips and completing multi-day scenarios in minutes. The modular design enables users to experiment with parameters, test algorithms, and integrate custom datasets, making the tool highly adaptable for diverse urban contexts. By providing a realistic and extensible platform, this adaptable, scalable, and open-source framework advances the optimization of EV fleet operations and offers a valuable resource for decision-makers and city planners navigating the transition to sustainable urban mobility solutions. We also present a case study using the NYC taxi dataset evaluating various dispatching algorithms, including closest vehicle dispatch, closest available vehicle dispatch, and power-of-d vehicle dispatch, and exploring charging approaches like continuous and nighttime charging. We propose a novel adaptive power-of-d dispatch policy, which dynamically adjusts to real-time conditions and demonstrates high throughputs when combined with adaptive charging policies that interrupt charging to meet demand during the peak and delay some of the charging to the nighttime.

math.OC↗

arXiv2024

Theseus: Exploring Efficient Wafer-Scale Chip Design for Large Language Models

The emergence of the large language model~(LLM) poses an exponential growth of demand for computation throughput, memory capacity, and communication bandwidth. Such a demand growth has significantly surpassed the improvement of corresponding chip designs. With the advancement of fabrication and integration technologies, designers have been developing Wafer-Scale Chips~(WSCs) to scale up and exploit the limits of computation density, memory capacity, and communication bandwidth at the level of a single chip. Existing solutions have demonstrated the significant advantages of WSCs over traditional designs, showing potential to effectively support LLM workloads. Despite the benefits, exploring the early-stage design space of WSCs for LLMs is a crucial yet challenging task due to the enormous and complicated design space, time-consuming evaluation methods, and inefficient exploration strategies. To address these challenges, we propose Theseus, an efficient WSC design space exploration framework for LLMs. We construct the design space of WSCs with various constraints considering the unique characteristics of WSCs. We propose efficient evaluation methodologies for large-scale NoC-based WSCs and introduce multi-fidelity Bayesian optimization to efficiently explore the design space. Evaluation results demonstrate the efficiency of Theseus that the searched Pareto optimal results outperform GPU cluster and existing WSC designs by up to 62.8\%/73.7\% in performance and 38.6\%/42.4\% in power consumption for LLM training, while improving up to 23.2$\times$ and 15.7$\times$ for the performance and power of inference tasks. Furthermore, we conduct case studies to address the design tradeoffs in WSCs and provide insights to facilitate WSC designs for LLMs.

cs.AR↗

arXiv2024

Quad-cascade picture of electrokinetic turbulence

Turbulence, ubiquitous in nature and across various systems, exhibits chaotic and intermittent fluctuations in space and time, defying precise prediction. For nearly a century, extensive efforts have been made to uncover the underlying universality and invariant laws from the immense disorder and chaotic nature of turbulence. While the celebrated Kolmogorov -5/3 law stands as a robust cornerstone, it falls short in capturing the diverse scaling behavior exhibited in turbulence influenced by external volume forces, like thermal convection and electrokinetic flows. This study proposes a general framework that couples the fluxes of kinetic energy and scalar variance, culminating in the formulation of a universal conservation law. This framework offers a comprehensive quad-cascade depiction of turbulence, enabling predictions that beyond the limitations of existing models. We illustrate this framework with microfluidic experiments on electrokinetic turbulence, wherein power spectra of concentration and velocity fluctuations exhibit the predicted scaling behaviors, providing remarkable agreement with theory. These findings not only deepen our understanding of the complete cascade process in turbulence driven by external volume forces but also hold promise for insights into other turbulent systems.

physics.flu-dyn↗