arXiv subjects

Nan Zhang

Publications and source records attributed to Nan Zhang.

At least 91 records · Page 5Linked to original sources

arXiv2025

Low temperature state in strontium titanate microcrystals using in situ multi-reflection Bragg coherent X-ray diffraction imaging

Strontium titanate is a classic quantum paraelectric oxide material that has been widely studied in bulk and thin films. It exhibits a well-known cubic-to-tetragonal antiferrodistortive phase transition at 105 K, characterized by the rotation of oxygen octahedra. A possible second phase transition at lower temperature is suppressed by quantum fluctuations, preventing the onset of ferroelectric order. However, recent studies have shown that ferroelectric order can be established at low temperatures by inducing strain and other means. Here, we used in situ multi-reflection Bragg coherent x-ray diffraction imaging to measure the strain and rotation tensors for two strontium titanate microcrystals at low temperature. We observe strains induced by dislocations and inclusion-like impurities in the microcrystals. Based on radial magnitude plots, these strains increase in magnitude and spread as the temperature decreases. Pearson's correlation heat maps show a structural transition at 50 K, which could possibly be the formation of a low-temperature ferroelectric phase in the presence of strain. We do not observe any change in local strains associated with the tetragonal phase transition at 105 K.

cond-mat.mtrl-sci↗

arXiv2025

A Foundational Generative Model for Breast Ultrasound Image Analysis

Foundational models have emerged as powerful tools for addressing various tasks in clinical settings. However, their potential development to breast ultrasound analysis remains untapped. In this paper, we present BUSGen, the first foundational generative model specifically designed for breast ultrasound image analysis. Pretrained on over 3.5 million breast ultrasound images, BUSGen has acquired extensive knowledge of breast structures, pathological features, and clinical variations. With few-shot adaptation, BUSGen can generate repositories of realistic and informative task-specific data, facilitating the development of models for a wide range of downstream tasks. Extensive experiments highlight BUSGen's exceptional adaptability, significantly exceeding real-data-trained foundational models in breast cancer screening, diagnosis, and prognosis. In breast cancer early diagnosis, our approach outperformed all board-certified radiologists (n=9), achieving an average sensitivity improvement of 16.5% (P-value<0.0001). Additionally, we characterized the scaling effect of using generated data which was as effective as the collected real-world data for training diagnostic models. Moreover, extensive experiments demonstrated that our approach improved the generalization ability of downstream models. Importantly, BUSGen protected patient privacy by enabling fully de-identified data sharing, making progress forward in secure medical data utilization. An online demo of BUSGen is available at https://aibus.bio.

cs.AI↗

arXiv2024

Beyond Efficiency: A Systematic Survey of Resource-Efficient Large Language Models

The burgeoning field of Large Language Models (LLMs), exemplified by sophisticated models like OpenAI's ChatGPT, represents a significant advancement in artificial intelligence. These models, however, bring forth substantial challenges in the high consumption of computational, memory, energy, and financial resources, especially in environments with limited resource capabilities. This survey aims to systematically address these challenges by reviewing a broad spectrum of techniques designed to enhance the resource efficiency of LLMs. We categorize methods based on their optimization focus: computational, memory, energy, financial, and network resources and their applicability across various stages of an LLM's lifecycle, including architecture design, pretraining, finetuning, and system design. Additionally, the survey introduces a nuanced categorization of resource efficiency techniques by their specific resource types, which uncovers the intricate relationships and mappings between various resources and corresponding optimization techniques. A standardized set of evaluation metrics and datasets is also presented to facilitate consistent and fair comparisons across different models and techniques. By offering a comprehensive overview of the current sota and identifying open research avenues, this survey serves as a foundational reference for researchers and practitioners, aiding them in developing more sustainable and efficient LLMs in a rapidly evolving landscape.

cs.LG↗

arXiv2024

Point Cloud Semantic Segmentation with Sparse and Inhomogeneous Annotations

Utilizing uniformly distributed sparse annotations, weakly supervised learning alleviates the heavy reliance on fine-grained annotations in point cloud semantic segmentation tasks. However, few works discuss the inhomogeneity of sparse annotations, albeit it is common in real-world scenarios. Therefore, this work introduces the probability density function into the gradient sampling approximation method to qualitatively analyze the impact of annotation sparsity and inhomogeneity under weakly supervised learning. Based on our analysis, we propose an Adaptive Annotation Distribution Network (AADNet) capable of robust learning on arbitrarily distributed sparse annotations. Specifically, we propose a label-aware point cloud downsampling strategy to increase the proportion of annotations involved in the training stage. Furthermore, we design the multiplicative dynamic entropy as the gradient calibration function to mitigate the gradient bias caused by non-uniformly distributed sparse annotations and explicitly reduce the epistemic uncertainty. Without any prior restrictions and additional information, our proposed method achieves comprehensive performance improvements at multiple label rates and different annotation distributions.

cs.CV↗

arXiv2024

Fermiology with nodal structures in nonsymmorphic superconductor LaNiGa$_2$: A de Haas-van Alphen study

Topological metals possess various types of symmetry-protected degenerate band crossings. When a topological metal becomes superconducting, the low-energy electronic excitations stemming from the band crossings located close to the Fermi level may contribute to highly unusual pairing symmetry and superconducting states. In this work, we study the electronic band structure of the time-reversal symmetry breaking superconductor LaNiGa$_2$ by means of quantum oscillation measurements. A comprehensive investigation combining angle-resolved high-field de Haas-van Alphen (dHvA) spectroscopy and first-principles calculations reveals the fermiology of LaNiGa$_2$ and verifies its nonsymmorphic $Cmcm$ lattice symmetry, which promises nodal band crossings pinned at the Fermi level with fourfold degeneracies. Moreover, such nodal structures, proposed to play a crucial role giving rise to the interorbital triplet pairing, are indeed captured by our dHvA analysis. Our results identify LaNiGa$_2$ as a prototypical topological crystalline superconductor and highlight the putative contribution of low-energy nodal quasiparticles to unconventional superconducting pairing.

cond-mat.supr-con↗

arXiv2024

Perturbation theory for resonant states near a bound state in the continuum

In this work, we develop a perturbation theory to analyze resonant states near a bound state in the continuum (BIC) in photonic crystal slabs. The theory allows us to rigorously determine the asymptotic behavior of $Q$-factor and the far-field polarization. We show that the resonant states close to a BIC can be nearly circularly polarized if the scattering matrix satisfies a certain condition. Moreover, our theory offers a novel perspective on super-BICs and provides a clear and precise condition to efficiently identify them in both symmetric and asymmetric structures without requiring a merging process. For practical applications, we find a super-BIC in a square lattice of rods on a dielectric substrate. Our theory addresses the non-Hermitian nature of the system, can be generalized to treat other structures that support BICs, and has potential applications in resonance and chiral optics.

physics.optics↗

arXiv2024

Dynamic Digital Twins of Blockchain Systems: State Extraction and Mirroring

Blockchain adoption is reaching an all-time high, with a plethora of blockchain architectures being developed to cover the needs of applications eager to integrate blockchain into their operations. However, blockchain systems suffer from the trilemma trade-off problem, which limits their ability to scale without sacrificing essential metrics such as decentralisation and security. The balance of the trilemma trade-off is primarily dictated by the consensus protocol used. Since consensus protocols are designed to function well under specific system conditions, and consequently, due to the blockchain's complex and dynamic nature, systems operating under a single consensus protocol are bound to face periods of inefficiency. The work presented in this paper constitutes part of an effort to design a Digital Twin-based blockchain management framework to balance the trilemma trade-off problem, which aims to adapt the consensus process to fit the conditions of the underlying system. Specifically, this work addresses the problems of extracting the blockchain system and mirroring it in its digital twin by proposing algorithms that overcome the challenges posed by blockchains' decentralised and asynchronous nature and the fundamental problems of global state and synchronisation in such systems. The robustness of the proposed algorithms is experimentally evaluated.

cs.CR↗

arXiv2024

When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs

Self-correction is an approach to improving responses from large language models (LLMs) by refining the responses using LLMs during inference. Prior work has proposed various self-correction frameworks using different sources of feedback, including self-evaluation and external feedback. However, there is still no consensus on the question of when LLMs can correct their own mistakes, as recent studies also report negative results. In this work, we critically survey broad papers and discuss the conditions required for successful self-correction. We first find that prior studies often do not define their research questions in detail and involve impractical frameworks or unfair evaluations that over-evaluate self-correction. To tackle these issues, we categorize research questions in self-correction research and provide a checklist for designing appropriate experiments. Our critical survey based on the newly categorized research questions shows that (1) no prior work demonstrates successful self-correction with feedback from prompted LLMs, except for studies in tasks that are exceptionally suited for self-correction, (2) self-correction works well in tasks that can use reliable external feedback, and (3) large-scale fine-tuning enables self-correction.

cs.CL↗

arXiv2024

GCLS$^2$: Towards Efficient Community Detection Using Graph Contrastive Learning with Structure Semantics

Due to the power of learning representations from unlabeled graphs, graph contrastive learning (GCL) has shown excellent performance in community detection tasks. Existing GCL-based methods on the community detection usually focused on learning attribute representations of individual nodes, which, however, ignores structural semantics of communities (e.g., nodes in the same community should be structurally cohesive). Therefore, in this paper, we will consider the community detection under the community structure semantics and propose an effective framework for graph contrastive learning under structure semantics (GCLS$^2$) to detect communities. To seamlessly integrate interior dense and exterior sparse characteristics of communities with our contrastive learning strategy, we employ classic community structures to extract high-level structural views and design a structure semantic expression module to augment the original structural feature representation. Moreover, we formulate the structure contrastive loss to optimize the feature representation of nodes, which can better capture the topology of communities. To adapt to large-scale networks, we design a high-level graph partitioning (HGP) algorithm that minimizes the community detection loss for GCLS$^2$ online training. It is worth noting that we prove a lower bound on the training of GCLS$^2$ from the perspective of the information theory, explaining why GCLS$^2$ can learn a more accurate representation of the structure. Extensive experiments have been conducted on various real-world graph datasets and confirmed that GCLS$^2$ outperforms nine state-of-the-art methods, in terms of the accuracy, modularity, and efficiency of detecting communities.

cs.SI↗

arXiv2024

Abnormally enhanced Hall Lorenz number in the magnetic Weyl semimetal NdAlSi

In Landau's celebrated Fermi liquid theory, electrons in a metal obey the Wiedemann--Franz law at the lowest temperatures. This law states that electron heat and charge transport are linked by a constant $L_0$, i.e., the Sommerfeld value of the Lorenz number ($L$). Such relation can be violated at elevated temperatures where the abundant inelastic scattering leads to a reduction of the Lorenz number ($L < L_0$). Here, we report a rare case of remarkably enhanced Lorenz number ($L > L_0$) discovered in the magnetic topological semimetal NdAlSi. Measurements of the transverse electrical and thermal transport coefficients reveal that the Hall Lorenz number $L_{xy}$ in NdAlSi starts to deviate from the canonical value far above its magnetic ordering temperature. Moreover, $L_{xy}$ displays strong nonmonotonic temperature and field dependence, reaching its maximum value close to 2$L_0$ in an intermediate parameter range. Further analysis excludes charge-neutral excitations as the origin of enhanced $L_{xy}$. Alternatively, we attribute it to the Kondo-type elastic scattering off localized 4$f$ electrons, which creates a peculiar energy distribution of the quasiparticle relaxation time. Our results provide insights into the perplexing transport phenomena caused by the interplay between charge and spin degrees of freedom.

cond-mat.str-el↗

arXiv2024

Two-dimensional superconductivity and anomalous vortex dissipation in newly-discovered transition metal dichalcogenide-based superlattices

Properties of layered superconductors can vary drastically when thinned down from bulk to monolayer, owing to the reduced dimensionality and weakened interlayer coupling. In transition metal dichalcogenides (TMDs), the inherent symmetry breaking effect in atomically thin crystals prompts novel states of matter, such as Ising superconductivity with an extraordinary in-plane upper critical field. Here, we demonstrate that two-dimensional (2D) superconductivity resembling those in atomic layers but with more fascinating behaviours can be realized in the bulk crystals of two new TMD-based superconductors Ba0.75ClTaS2 and Ba0.75ClTaSe2. They comprise an alternating stack of H-type TMD layers and Ba-Cl layers. In both materials, intrinsic 2D superconductivity develops below a Berezinskii-Kosterlitz-Thouless transition. The upper critical field along ab plane exceeds the Pauli limit (Hp); in particular, Ba0.75ClTaSe2 exhibits an extremely high in plane Hc2 (14Hp) and a colossal superconducting anisotropy of 150. Moreover, the temperature-field phase diagram of Ba0.75ClTaSe2 under an in-plane magnetic field contains a large phase regime of vortex dissipation, which can be ascribed to the Josephson vortex motion, signifying an unprecedentedly strong fluctuation effect in TMD-based superconductors. Our results provide a new path towards the establishment of 2D superconductivity and novel exotic quantum phases in bulk crystals of TMD-based superconductors.

cond-mat.supr-con↗

arXiv2024

How much does AI impact development speed? An enterprise-based randomized controlled trial

How much does AI assistance impact developer productivity? To date, the software engineering literature has provided a range of answers, targeting a diversity of outcomes: from perceived productivity to speed on task and developer throughput. Our randomized controlled trial with 96 full-time Google software engineers contributes to this literature by sharing an estimate of the impact of three AI features on the time developers spent on a complex, enterprise-grade task. We found that AI significantly shortened the time developers spent on task. Our best estimate of the size of this effect, controlling for factors known to influence developer time on task, stands at about 21\%, although our confidence interval is large. We also found an interesting effect whereby developers who spend more hours on code-related activities per day were faster with AI. Product and future research considerations are discussed. In particular, we invite further research that explores the impact of AI at the ecosystem level and across multiple suites of AI-enhanced tools, since we cannot assume that the effect size obtained in our lab study will necessarily apply more broadly, or that the effect of AI found using internal Google tooling in the summer of 2024 will translate across tools and over time.

cs.SE↗

arXiv2024

Knowledge Equivalence in Digital Twins of Intelligent Systems

A digital twin contains up-to-date data-driven models of the physical world being studied and can use simulation to optimise the physical world. However, the analysis made by the digital twin is valid and reliable only when the model is equivalent to the physical world. Maintaining such an equivalent model is challenging, especially when the physical systems being modelled are intelligent and autonomous. The paper focuses in particular on digital twin models of intelligent systems where the systems are knowledge-aware but with limited capability. The digital twin improves the acting of the physical system at a meta-level by accumulating more knowledge in the simulated environment. The modelling of such an intelligent physical system requires replicating the knowledge-awareness capability in the virtual space. Novel equivalence maintaining techniques are needed, especially in synchronising the knowledge between the model and the physical system. This paper proposes the notion of knowledge equivalence and an equivalence maintaining approach by knowledge comparison and updates. A quantitative analysis of the proposed approach confirms that compared to state equivalence, knowledge equivalence maintenance can tolerate deviation thus reducing unnecessary updates and achieve more Pareto efficient solutions for the trade-off between update overhead and simulation reliability.

cs.AI↗

arXiv2024

LLMs Assist NLP Researchers: Critique Paper (Meta-)Reviewing

This work is motivated by two key trends. On one hand, large language models (LLMs) have shown remarkable versatility in various generative tasks such as writing, drawing, and question answering, significantly reducing the time required for many routine tasks. On the other hand, researchers, whose work is not only time-consuming but also highly expertise-demanding, face increasing challenges as they have to spend more time reading, writing, and reviewing papers. This raises the question: how can LLMs potentially assist researchers in alleviating their heavy workload? This study focuses on the topic of LLMs assist NLP Researchers, particularly examining the effectiveness of LLM in assisting paper (meta-)reviewing and its recognizability. To address this, we constructed the ReviewCritique dataset, which includes two types of information: (i) NLP papers (initial submissions rather than camera-ready) with both human-written and LLM-generated reviews, and (ii) each review comes with "deficiency" labels and corresponding explanations for individual segments, annotated by experts. Using ReviewCritique, this study explores two threads of research questions: (i) "LLMs as Reviewers", how do reviews generated by LLMs compare with those written by humans in terms of quality and distinguishability? (ii) "LLMs as Metareviewers", how effectively can LLMs identify potential issues, such as Deficient or unprofessional review segments, within individual paper reviews? To our knowledge, this is the first work to provide such a comprehensive analysis.

cs.CL↗

arXiv2024

From MOOC to MAIC: Reshaping Online Teaching and Learning through LLM-driven Agents

Since the first instances of online education, where courses were uploaded to accessible and shared online platforms, this form of scaling the dissemination of human knowledge to reach a broader audience has sparked extensive discussion and widespread adoption. Recognizing that personalized learning still holds significant potential for improvement, new AI technologies have been continuously integrated into this learning format, resulting in a variety of educational AI applications such as educational recommendation and intelligent tutoring. The emergence of intelligence in large language models (LLMs) has allowed for these educational enhancements to be built upon a unified foundational model, enabling deeper integration. In this context, we propose MAIC (Massive AI-empowered Course), a new form of online education that leverages LLM-driven multi-agent systems to construct an AI-augmented classroom, balancing scalability with adaptivity. Beyond exploring the conceptual framework and technical innovations, we conduct preliminary experiments at Tsinghua University, one of China's leading universities. Drawing from over 100,000 learning records of more than 500 students, we obtain a series of valuable observations and initial analyses. This project will continue to evolve, ultimately aiming to establish a comprehensive open platform that supports and unifies research, technology, and applications in exploring the possibilities of online education in the era of large model AI. We envision this platform as a collaborative hub, bringing together educators, researchers, and innovators to collectively explore the future of AI-driven online education.

cs.CY↗

arXiv2024

Large Language Models for Explainable Decisions in Dynamic Digital Twins

Dynamic data-driven Digital Twins (DDTs) can enable informed decision-making and provide an optimisation platform for the underlying system. By leveraging principles of Dynamic Data-Driven Applications Systems (DDDAS), DDTs can formulate computational modalities for feedback loops, model updates and decision-making, including autonomous ones. However, understanding autonomous decision-making often requires technical and domain-specific knowledge. This paper explores using large language models (LLMs) to provide an explainability platform for DDTs, generating natural language explanations of the system's decision-making by leveraging domain-specific knowledge bases. A case study from smart agriculture is presented.

cs.AI↗

arXiv2024

Reverse Influential Community Search Over Social Networks (Technical Report)

As an important fundamental task of numerous real-world applications such as social network analysis and online advertising/marketing, several prior works studied influential community search, which retrieves a community with high structural cohesiveness and maximum influences on other users in social networks. However, previous works usually considered the influences of the community on arbitrary users in social networks, rather than specific groups (e.g., customer groups, or senior communities). Inspired by this, we propose a novel Top-M Reverse Influential Community Search (TopM-RICS) problem, which obtains a seed community with the maximum influence on a user-specified target community, satisfying both structural and keyword constraints. To efficiently tackle the TopM-RICS problem, we design effective pruning strategies to filter out false alarms of candidate seed communities, and propose an effective index mechanism to facilitate the community retrieval. We also formulate and tackle a TopM-RICS variant, named Top-M Relaxed Reverse Influential Community Search} (TopM-R2ICS), which returns top-M subgraphs with relaxed structural constraints and having the maximum influence on a user-specified target community. Comprehensive experiments have been conducted to verify the efficiency and effectiveness of our TopM-RICS and TopM-R2ICS approaches on both real-world and synthetic social networks under various parameter settings.

cs.SI↗

arXiv2024

Evaluating LLMs at Detecting Errors in LLM Responses

With Large Language Models (LLMs) being widely used across various tasks, detecting errors in their responses is increasingly crucial. However, little research has been conducted on error detection of LLM responses. Collecting error annotations on LLM responses is challenging due to the subjective nature of many NLP tasks, and thus previous research focuses on tasks of little practical value (e.g., word sorting) or limited error types (e.g., faithfulness in summarization). This work introduces ReaLMistake, the first error detection benchmark consisting of objective, realistic, and diverse errors made by LLMs. ReaLMistake contains three challenging and meaningful tasks that introduce objectively assessable errors in four categories (reasoning correctness, instruction-following, context-faithfulness, and parameterized knowledge), eliciting naturally observed and diverse errors in responses of GPT-4 and Llama 2 70B annotated by experts. We use ReaLMistake to evaluate error detectors based on 12 LLMs. Our findings show: 1) Top LLMs like GPT-4 and Claude 3 detect errors made by LLMs at very low recall, and all LLM-based error detectors perform much worse than humans. 2) Explanations by LLM-based error detectors lack reliability. 3) LLMs-based error detection is sensitive to small changes in prompts but remains challenging to improve. 4) Popular approaches to improving LLMs, including self-consistency and majority vote, do not improve the error detection performance. Our benchmark and code are provided at https://github.com/psunlpgroup/ReaLMistake.

cs.CL↗