arXiv subjects

Yu Wang

Publications and source records attributed to Yu Wang.

At least 811 records · Page 45Linked to original sources

arXiv2024

Can AI Understand Our Universe? Test of Fine-Tuning GPT by Astrophysical Data

ChatGPT has been the most talked-about concept in recent months, captivating both professionals and the general public alike, and has sparked discussions about the changes that artificial intelligence (AI) will bring to the world. As physicists and astrophysicists, we are curious about if scientific data can be correctly analyzed by large language models (LLMs) and yield accurate physics. In this article, we fine-tune the generative pre-trained transformer (GPT) model by the astronomical data from the observations of galaxies, quasars, stars, gamma-ray bursts (GRBs), and the simulations of black holes (BHs), the fine-tuned model demonstrates its capability to classify astrophysical phenomena, distinguish between two types of GRBs, deduce the redshift of quasars, and estimate BH parameters. We regard this as a successful test, marking the LLM's proven efficacy in scientific research. With the ever-growing volume of multidisciplinary data and the advancement of AI technology, we look forward to the emergence of a more fundamental and comprehensive understanding of our universe. This article also shares some interesting thoughts on data collection and AI design. Using the approach of understanding the universe - looking outward at data and inward for fundamental building blocks - as a guideline, we propose a method of series expansion for AI, suggesting ways to train and control AI that is smarter than humans.

astro-ph.IM↗

arXiv2024

MING-MOE: Enhancing Medical Multi-Task Learning in Large Language Models with Sparse Mixture of Low-Rank Adapter Experts

Large language models like ChatGPT have shown substantial progress in natural language understanding and generation, proving valuable across various disciplines, including the medical field. Despite advancements, challenges persist due to the complexity and diversity inherent in medical tasks which often require multi-task learning capabilities. Previous approaches, although beneficial, fall short in real-world applications because they necessitate task-specific annotations at inference time, limiting broader generalization. This paper introduces MING-MOE, a novel Mixture-of-Expert~(MOE)-based medical large language model designed to manage diverse and complex medical tasks without requiring task-specific annotations, thus enhancing its usability across extensive datasets. MING-MOE employs a Mixture of Low-Rank Adaptation (MoLoRA) technique, allowing for efficient parameter usage by maintaining base model parameters static while adapting through a minimal set of trainable parameters. We demonstrate that MING-MOE achieves state-of-the-art (SOTA) performance on over 20 medical tasks, illustrating a significant improvement over existing models. This approach not only extends the capabilities of medical language models but also improves inference efficiency.

cs.CL↗

arXiv2024

Null controllability for stochastic coupled systems of fourth order parabolic equations

This paper aims to establish null controllability for systems coupled by two backward fourth order stochastic parabolic equations. The main goal is to control both equations with only one control act on the drift term. To achieve this, we develop a new global Carleman estimate for fourth order stochastic parabolic equations, which allows us to deduce a suitable observability inequality for the adjoint systems.

math.OC↗

arXiv2024

Floquet non-equilibrium Green's function and Floquet quantum master equation for electronic transport: The role of electron-electron interactions and spin current with circular light

Non-equilibrium Green's function (NEGF) and quantum master equation (QME) are two main classes of approaches for electronic transport. We discuss various Floquet variances of these formalisms for transport properties of a quantum dot driven via interaction with an external periodic field. We first derived two versions of the Floquet NEGF. We also explore an ansatz of the Floquet NEGF formalism for the interacting systems. In addition, we derived two versions of Floquet QME in the weak interaction regime. With each method, we elaborate on the evaluation of the expectation values of the number and current operators. We examined these methods for transport through a two-level system that is subject to periodic driving. The numerical results of all four methods show good agreement for non-interacting systems in the weak regime. Furthermore, we have observed that circular light can introduce spin current. We expect these Floquet quantum transport methods to be useful in studying molecular junctions exposed to light.

quant-ph↗

arXiv2024

On the Uniqueness of Solution for the Bellman Equation of LTL Objectives

Surrogate rewards for linear temporal logic (LTL) objectives are commonly utilized in planning problems for LTL objectives. In a widely-adopted surrogate reward approach, two discount factors are used to ensure that the expected return approximates the satisfaction probability of the LTL objective. The expected return then can be estimated by methods using the Bellman updates such as reinforcement learning. However, the uniqueness of the solution to the Bellman equation with two discount factors has not been explicitly discussed. We demonstrate with an example that when one of the discount factors is set to one, as allowed in many previous works, the Bellman equation may have multiple solutions, leading to inaccurate evaluation of the expected return. We then propose a condition for the Bellman equation to have the expected return as the unique solution, requiring the solutions for states inside a rejecting bottom strongly connected component (BSCC) to be 0. We prove this condition is sufficient by showing that the solutions for the states with discounting can be separated from those for the states without discounting under this condition

cs.AI↗

arXiv2024

Understanding Energy Level Structure Using Quantum Rubik's Cube

This study combines the quantum Rubik's Cube matrix with the Benalcazar Bernevig Hughes model, defines a matrix algorithm based on the reverse process of convolution, and constructs an expression for the quantum Rubik's Cube matrix and Hamiltonian. Furthermore, in order to make the operation of the quantum Rubik's Cube matrix clearer, we use a Josephus ring to draw a topological graph of the Rubik's Cube expansion. This article uses a quantum Rubik's Cube to calculate energy level transitions of electrons, and shows that its operation corresponds to path integration. The band dispersion is obtained. This work provides new ideas and methods for calculating Hamiltonians and studying energy level structure.

quant-ph↗

arXiv2024

LLMs in Political Science: Heralding a New Era of Visual Analysis

Interest is increasing among political scientists in leveraging the extensive information available in images. However, the challenge of interpreting these images lies in the need for specialized knowledge in computer vision and access to specialized hardware. As a result, image analysis has been limited to a relatively small group within the political science community. This landscape could potentially change thanks to the rise of large language models (LLMs). This paper aims to raise awareness of the feasibility of using Gemini for image content analysis. A retrospective analysis was conducted on a corpus of 688 images. Content reports were elicited from Gemini for each image and then manually evaluated by the authors. We find that Gemini is highly accurate in performing object detection, which is arguably the most common and fundamental task in image analysis for political scientists. Equally important, we show that it is easy to implement as the entire command consists of a single prompt in natural language; it is fast to run and should meet the time budget of most researchers; and it is free to use and does not require any specialized hardware. In addition, we illustrate how political scientists can leverage Gemini for other image understanding tasks, including face identification, sentiment analysis, and caption generation. Our findings suggest that Gemini and other similar LLMs have the potential to drastically stimulate and accelerate image research in political science and social sciences more broadly.

cs.CV↗

arXiv2024

Leveraging Large Language Models for Fuzzy String Matching in Political Science

Fuzzy string matching remains a key issue when political scientists combine data from different sources. Existing matching methods invariably rely on string distances, such as Levenshtein distance and cosine similarity. As such, they are inherently incapable of matching strings that refer to the same entity with different names such as ''JP Morgan'' and ''Chase Bank'', ''DPRK'' and ''North Korea'', ''Chuck Fleischmann (R)'' and ''Charles Fleischmann (R)''. In this letter, we propose to use large language models to entirely sidestep this problem in an easy and intuitive manner. Extensive experiments show that our proposed methods can improve the state of the art by as much as 39% in terms of average precision while being substantially easier and more intuitive to use by political scientists. Moreover, our results are robust against various temperatures. We further note that enhanced prompting can lead to additional performance improvements.

cs.AI↗

arXiv2024

Phase transition in the EM scheme of an SDE driven by $α$-stable noises with $α\in (0,2]$

We study in this paper the EM scheme for a family of well-posed critical SDEs with the drift $-x\log(1+|x|)$ and $α$-stable noises. Specifically, we find that when the SDE is driven by a rotationally symmetric $α$-stable processes with $α=2$ (i.e. Brownian motion), the EM scheme is bounded in the $L^2$ sense uniformly w.r.t. the time. In contrast, if the SDE is driven by a rotationally symmetric $α$-stable process with $α\in (0,2)$, all the $β$-th moments, with $β\in (0,α)$, of the EM scheme blow up. This demonstrates a phase transition phenomenon as $α\uparrow 2$. We verify our results by simulations.

math.PR↗

arXiv2024

Pushing Auto-regressive Models for 3D Shape Generation at Capacity and Scalability

Auto-regressive models have achieved impressive results in 2D image generation by modeling joint distributions in grid space. In this paper, we extend auto-regressive models to 3D domains, and seek a stronger ability of 3D shape generation by improving auto-regressive models at capacity and scalability simultaneously. Firstly, we leverage an ensemble of publicly available 3D datasets to facilitate the training of large-scale models. It consists of a comprehensive collection of approximately 900,000 objects, with multiple properties of meshes, points, voxels, rendered images, and text captions. This diverse labeled dataset, termed Objaverse-Mix, empowers our model to learn from a wide range of object variations. However, directly applying 3D auto-regression encounters critical challenges of high computational demands on volumetric grids and ambiguous auto-regressive order along grid dimensions, resulting in inferior quality of 3D shapes. To this end, we then present a novel framework Argus3D in terms of capacity. Concretely, our approach introduces discrete representation learning based on a latent vector instead of volumetric grids, which not only reduces computational costs but also preserves essential geometric details by learning the joint distributions in a more tractable order. The capacity of conditional generation can thus be realized by simply concatenating various conditioning inputs to the latent vector, such as point clouds, categories, images, and texts. In addition, thanks to the simplicity of our model architecture, we naturally scale up our approach to a larger model with an impressive 3.6 billion parameters, further enhancing the quality of versatile 3D generation. Extensive experiments on four generation tasks demonstrate that Argus3D can synthesize diverse and faithful shapes across multiple categories, achieving remarkable performance.

cs.CV↗

arXiv2024

FlashEval: Towards Fast and Accurate Evaluation of Text-to-image Diffusion Generative Models

In recent years, there has been significant progress in the development of text-to-image generative models. Evaluating the quality of the generative models is one essential step in the development process. Unfortunately, the evaluation process could consume a significant amount of computational resources, making the required periodic evaluation of model performance (e.g., monitoring training progress) impractical. Therefore, we seek to improve the evaluation efficiency by selecting the representative subset of the text-image dataset. We systematically investigate the design choices, including the selection criteria (textural features or image-based metrics) and the selection granularity (prompt-level or set-level). We find that the insights from prior work on subset selection for training data do not generalize to this problem, and we propose FlashEval, an iterative search algorithm tailored to evaluation data selection. We demonstrate the effectiveness of FlashEval on ranking diffusion models with various configurations, including architectures, quantization levels, and sampler schedules on COCO and DiffusionDB datasets. Our searched 50-item subset could achieve comparable evaluation quality to the randomly sampled 500-item subset for COCO annotations on unseen models, achieving a 10x evaluation speedup. We release the condensed subset of these commonly used datasets to help facilitate diffusion algorithm design and evaluation, and open-source FlashEval as a tool for condensing future datasets, accessible at https://github.com/thu-nics/FlashEval.

cs.CV↗

arXiv2024

Multi-view Deep Subspace Clustering Networks

Multi-view subspace clustering aims to discover the inherent structure of data by fusing multiple views of complementary information. Most existing methods first extract multiple types of handcrafted features and then learn a joint affinity matrix for clustering. The disadvantage of this approach lies in two aspects: 1) multi-view relations are not embedded into feature learning, and 2) the end-to-end learning manner of deep learning is not suitable for multi-view clustering. Even when deep features have been extracted, it is a nontrivial problem to choose a proper backbone for clustering on different datasets. To address these issues, we propose the Multi-view Deep Subspace Clustering Networks (MvDSCN), which learns a multi-view self-representation matrix in an end-to-end manner. The MvDSCN consists of two sub-networks, \ie, a diversity network (Dnet) and a universality network (Unet). A latent space is built using deep convolutional autoencoders, and a self-representation matrix is learned in the latent space using a fully connected layer. Dnet learns view-specific self-representation matrices, whereas Unet learns a common self-representation matrix for all views. To exploit the complementarity of multi-view representations, the Hilbert--Schmidt independence criterion (HSIC) is introduced as a diversity regularizer that captures the nonlinear, high-order inter-view relations. Because different views share the same label space, the self-representation matrices of each view are aligned to the common one by universality regularization. The MvDSCN also unifies multiple backbones to boost clustering performance and avoid the need for model selection. Experiments demonstrate the superiority of the MvDSCN.

cs.CV↗

arXiv2024

Mott insulating states of the anisotropic SU(4) Dirac fermions

We employ the large-scale quantum Monte-Carlo simulations to investigate the Mott-insulating states of the half-filled SU(4) Hubbard model on the square lattice with a staggered-flux pattern. The noninteracting band structure that evolves from a nested Fermi surface at zero flux to isotropic Dirac cones at $π$ flux, exhibits anisotropic Dirac cones as the flux varies in between. Our simulations show transitions between the three phases of Dirac semimetal, antiferromagnet and valence-bond solid. A direct continuous transition between the antiferromagnetic phase and the valence-bond-solid phase is realized via varying the flux in the Mott regime. The simulated critical exponents remarkably agree with those of SU(4) $J$-$Q$ model. Inside the valence-bond-solid phase induced by the flux, the plaquette valence-bond state with vanishing single-particle gap is identified. At strong coupling, the valence-bond-solid phase disappears and the Mott-insulating state is always accompanied by antiferromagnetic ordering, regardless of the magnitude of the flux.

cond-mat.quant-gas↗

arXiv2024

PrPSeg: Universal Proposition Learning for Panoramic Renal Pathology Segmentation

Understanding the anatomy of renal pathology is crucial for advancing disease diagnostics, treatment evaluation, and clinical research. The complex kidney system comprises various components across multiple levels, including regions (cortex, medulla), functional units (glomeruli, tubules), and cells (podocytes, mesangial cells in glomerulus). Prior studies have predominantly overlooked the intricate spatial interrelations among objects from clinical knowledge. In this research, we introduce a novel universal proposition learning approach, called panoramic renal pathology segmentation (PrPSeg), designed to segment comprehensively panoramic structures within kidney by integrating extensive knowledge of kidney anatomy. In this paper, we propose (1) the design of a comprehensive universal proposition matrix for renal pathology, facilitating the incorporation of classification and spatial relationships into the segmentation process; (2) a token-based dynamic head single network architecture, with the improvement of the partial label image segmentation and capability for future data enlargement; and (3) an anatomy loss function, quantifying the inter-object relationships across the kidney.

eess.IV↗

arXiv2024

Tokensome: Towards a Genetic Vision-Language GPT for Explainable and Cognitive Karyotyping

Automatic karyotype analysis is often defined as a visual perception task focused solely on chromosomal object-level modeling. This definition has led most existing methods to overlook componential and holistic information, significantly constraining model performance. Moreover, the lack of interpretability in current technologies hinders clinical adoption. In this paper, we introduce Tokensome, a novel vision-language model based on chromosome tokenization for explainable and cognitive karyotyping. Tokensome elevates the method from the conventional visual perception layer to the cognitive decision-making layer. This elevation enables the integration of domain knowledge and cognitive reasoning via knowledge graphs and LLMs, markedly enhancing model's explainability and facilitating abnormality detection.

cs.CV↗

arXiv2024

AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shield Prompting

With the advent and widespread deployment of Multimodal Large Language Models (MLLMs), the imperative to ensure their safety has become increasingly pronounced. However, with the integration of additional modalities, MLLMs are exposed to new vulnerabilities, rendering them prone to structured-based jailbreak attacks, where semantic content (e.g., "harmful text") has been injected into the images to mislead MLLMs. In this work, we aim to defend against such threats. Specifically, we propose \textbf{Ada}ptive \textbf{Shield} Prompting (\textbf{AdaShield}), which prepends inputs with defense prompts to defend MLLMs against structure-based jailbreak attacks without fine-tuning MLLMs or training additional modules (e.g., post-stage content detector). Initially, we present a manually designed static defense prompt, which thoroughly examines the image and instruction content step by step and specifies response methods to malicious queries. Furthermore, we introduce an adaptive auto-refinement framework, consisting of a target MLLM and a LLM-based defense prompt generator (Defender). These components collaboratively and iteratively communicate to generate a defense prompt. Extensive experiments on the popular structure-based jailbreak attacks and benign datasets show that our methods can consistently improve MLLMs' robustness against structure-based jailbreak attacks without compromising the model's general capabilities evaluated on standard benign tasks. Our code is available at https://github.com/rain305f/AdaShield.

cs.CR↗

arXiv2024

Localization matters too: How localization error affects UAV flight

The maximum safe flight speed of a Unmanned Aerial Vehicle (UAV) is an important indicator for measuring its efficiency in completing various tasks. This indicator is influenced by numerous parameters such as UAV localization error, perception range, and system latency. However, in terms of localization errors, although there have been many studies dedicated to improving the localization capability of UAVs, there is a lack of quantitative research on their impact on speed. In this work, we model the relationship between various parameters of the UAV and its maximum flight speed. We consider a scenario similar to navigating through dense forests, where the UAV needs to quickly avoid obstacles directly ahead and swiftly reorient after avoidance. Based on this scenario, we studied how parameters such as localization error affect the maximum safe speed during UAV flight, as well as the coupling relationships between these parameters. Furthermore, we validated our model in a simulation environment, and the results showed that the predicted maximum safe speed had an error of less than 20% compared to the test speed. In high-density situations, localization error has a significant impact on the UAV's maximum safe flight speed. This model can help designers utilize more suitable software and hardware to construct a UAV system.

cs.RO↗

arXiv2024

COLA: Cross-city Mobility Transformer for Human Trajectory Simulation

Human trajectory data produced by daily mobile devices has proven its usefulness in various substantial fields such as urban planning and epidemic prevention. In terms of the individual privacy concern, human trajectory simulation has attracted increasing attention from researchers, targeting at offering numerous realistic mobility data for downstream tasks. Nevertheless, the prevalent issue of data scarcity undoubtedly degrades the reliability of existing deep learning models. In this paper, we are motivated to explore the intriguing problem of mobility transfer across cities, grasping the universal patterns of human trajectories to augment the powerful Transformer with external mobility data. There are two crucial challenges arising in the knowledge transfer across cities: 1) how to transfer the Transformer to adapt for domain heterogeneity; 2) how to calibrate the Transformer to adapt for subtly different long-tail frequency distributions of locations. To address these challenges, we have tailored a Cross-city mObiLity trAnsformer (COLA) with a dedicated model-agnostic transfer framework by effectively transferring cross-city knowledge for human trajectory simulation. Firstly, COLA divides the Transformer into the private modules for city-specific characteristics and the shared modules for city-universal mobility patterns. Secondly, COLA leverages a lightweight yet effective post-hoc adjustment strategy for trajectory simulation, without disturbing the complex bi-level optimization of model-agnostic knowledge transfer. Extensive experiments of COLA compared to state-of-the-art single-city baselines and our implemented cross-city baselines have demonstrated its superiority and effectiveness. The code is available at https://github.com/Star607/Cross-city-Mobility-Transformer.

cs.LG↗