arXiv subjects

Yi Zhang

Publications and source records attributed to Yi Zhang.

At least 757 records · Page 42Linked to original sources

arXiv2023

UQ at #SMM4H 2023: ALEX for Public Health Analysis with Social Media

As social media becomes increasingly popular, more and more activities related to public health emerge. Current techniques for public health analysis involve popular models such as BERT and large language models (LLMs). However, the costs of training in-domain LLMs for public health are especially expensive. Furthermore, such kinds of in-domain datasets from social media are generally imbalanced. To tackle these challenges, the data imbalance issue can be overcome by data augmentation and balanced training. Moreover, the ability of the LLMs can be effectively utilized by prompting the model properly. In this paper, a novel ALEX framework is proposed to improve the performance of public health analysis on social media by adopting an LLMs explanation mechanism. Results show that our ALEX model got the best performance among all submissions in both Task 2 and Task 4 with a high score in Task 1 in Social Media Mining for Health 2023 (SMM4H)[1]. Our code has been released at https:// github.com/YanJiangJerry/ALEX.

cs.CL↗

arXiv2023

Balanced and Explainable Social Media Analysis for Public Health with Large Language Models

As social media becomes increasingly popular, more and more public health activities emerge, which is worth noting for pandemic monitoring and government decision-making. Current techniques for public health analysis involve popular models such as BERT and large language models (LLMs). Although recent progress in LLMs has shown a strong ability to comprehend knowledge by being fine-tuned on specific domain datasets, the costs of training an in-domain LLM for every specific public health task are especially expensive. Furthermore, such kinds of in-domain datasets from social media are generally highly imbalanced, which will hinder the efficiency of LLMs tuning. To tackle these challenges, the data imbalance issue can be overcome by sophisticated data augmentation methods for social media datasets. In addition, the ability of the LLMs can be effectively utilised by prompting the model properly. In light of the above discussion, in this paper, a novel ALEX framework is proposed for social media analysis on public health. Specifically, an augmentation pipeline is developed to resolve the data imbalance issue. Furthermore, an LLMs explanation mechanism is proposed by prompting an LLM with the predicted results from BERT models. Extensive experiments conducted on three tasks at the Social Media Mining for Health 2023 (SMM4H) competition with the first ranking in two tasks demonstrate the superior performance of the proposed ALEX method. Our code has been released in https://github.com/YanJiangJerry/ALEX.

cs.CL↗

arXiv2023

Transient analysis of arm locking controller

Arm locking is one of the key technologies to suppress the laser phase noise in spaced-based gravitational waves observatories. Since arm locking was proposed, phase margin criterion was always used as the fundamental design strategy for the controller development. In this paper, we find that this empirical method from engineering actually cannot guarantee the arm locking stability. Therefore, most of the advanced arm locking controllers reported so far may have instable problems. After comprehensive analysis of the single arm locking's transient responses, strict analytical stability criterions are summarized for the first time. These criterions are then generalized to dual arm locking, modified-dual arm locking and common arm locking, and special considerations for the design of arm locking controllers in different architectures are also discussed. It is found that PI controllers can easily meet our stability criterions in most of the arm locking systems. Using a simple high gain PI controller, it is possible to suppress the laser phase noise by 5 orders of magnitude within the science band. Our stability criterions can also be used in other feedback systems, where several modules with different delays are connected in parallel.

gr-qc↗

arXiv2023

Crystal-Chemical Origins of the Ultrahigh Conductivity of Metallic Delafossites

Despite their highly anisotropic complex-oxidic nature, certain delafossite compounds (e.g., PdCoO2, PtCoO2) are the most conductive oxides known, for reasons that remain poorly understood. Their room-temperature conductivity can exceed that of Au, while their low-temperature electronic mean-free-paths reach an astonishing 20 microns. It is widely accepted that these materials must be ultrapure to achieve this, although the methods for their growth (which produce only small crystals) are not typically capable of such. Here, we first report a new approach to PdCoO2 crystal growth, using chemical vapor transport methods to achieve order-of-magnitude gains in size, the highest structural qualities yet reported, and record residual resistivity ratios (>440). Nevertheless, the first detailed mass spectrometry measurements on these materials reveal that they are not ultrapure, typically harboring 100s-of-parts-per-million impurity levels. Through quantitative crystal-chemical analyses, we resolve this apparent dichotomy, showing that the vast majority of impurities are forced to reside in the Co-O octahedral layers, leaving the conductive Pd sheets highly pure (~1 ppm impurity concentrations). These purities are shown to be in quantitative agreement with measured residual resistivities. We thus conclude that a previously unconsidered "sublattice purification" mechanism is essential to the ultrahigh low-temperature conductivity and mean-free-path of metallic delafossites.

cond-mat.mtrl-sci↗

arXiv2023

BDC-Adapter: Brownian Distance Covariance for Better Vision-Language Reasoning

Large-scale pre-trained Vision-Language Models (VLMs), such as CLIP and ALIGN, have introduced a new paradigm for learning transferable visual representations. Recently, there has been a surge of interest among researchers in developing lightweight fine-tuning techniques to adapt these models to downstream visual tasks. We recognize that current state-of-the-art fine-tuning methods, such as Tip-Adapter, simply consider the covariance between the query image feature and features of support few-shot training samples, which only captures linear relations and potentially instigates a deceptive perception of independence. To address this issue, in this work, we innovatively introduce Brownian Distance Covariance (BDC) to the field of vision-language reasoning. The BDC metric can model all possible relations, providing a robust metric for measuring feature dependence. Based on this, we present a novel method called BDC-Adapter, which integrates BDC prototype similarity reasoning and multi-modal reasoning network prediction to perform classification tasks. Our extensive experimental results show that the proposed BDC-Adapter can freely handle non-linear relations and fully characterize independence, outperforming the current state-of-the-art methods by large margins.

cs.CV↗

arXiv2023

The Existence of Hamilton Cycle in n-Balanced k-Partite Graphs

Let $G_{k,n}$ be the $n$-balanced $k$-partite graph, whose vertex set can be partitioned into $k$ parts, each has $n$ vertices. In this paper, we prove that if $k \geq 2,n \geq 1$, for the edge set $E(G)$ of $G_{k,n}$ $$|E(G)| \geq\left\{\begin{array}{cc} 1 & \text { if } k=2, n=1 n^{2} C_{k}^{2}-(k-1) n+2 & \text { other } \end{array}\right.$$ then $G_{k,n}$ is hamiltonian. And the result may be the best.

math.CO↗

arXiv2023

Twisted Mahler discrete residues

Recently we constructed Mahler discrete residues for rational functions and showed they comprise a complete obstruction to the Mahler summability problem of deciding whether a given rational function $f(x)$ is of the form $g(x^p)-g(x)$ for some rational function $g(x)$ and an integer $p > 1$. Here we develop a notion of $λ$-twisted Mahler discrete residues for $λ\in\mathbb{Z}$, and show that they similarly comprise a complete obstruction to the twisted Mahler summability problem of deciding whether a given rational function $f(x)$ is of the form $p^λg(x^p)-g(x)$ for some rational function $g(x)$ and an integer $p>1$. We provide some initial applications of twisted Mahler discrete residues to differential creative telescoping problems for Mahler functions and to the differential Galois theory of linear Mahler equations.

math.NT↗

arXiv2023

Unsupervised Prototype Adapter for Vision-Language Models

Recently, large-scale pre-trained vision-language models (e.g. CLIP and ALIGN) have demonstrated remarkable effectiveness in acquiring transferable visual representations. To leverage the valuable knowledge encoded within these models for downstream tasks, several fine-tuning approaches, including prompt tuning methods and adapter-based methods, have been developed to adapt vision-language models effectively with supervision. However, these methods rely on the availability of annotated samples, which can be labor-intensive and time-consuming to acquire, thus limiting scalability. To address this issue, in this work, we design an unsupervised fine-tuning approach for vision-language models called Unsupervised Prototype Adapter (UP-Adapter). Specifically, for the unannotated target datasets, we leverage the text-image aligning capability of CLIP to automatically select the most confident samples for each class. Utilizing these selected samples, we generate class prototypes, which serve as the initialization for the learnable prototype model. After fine-tuning, the prototype model prediction is combined with the original CLIP's prediction by a residual connection to perform downstream recognition tasks. Our extensive experimental results on image recognition and domain generalization show that the proposed unsupervised method outperforms 8-shot CoOp, 8-shot Tip-Adapter, and also the state-of-the-art UPL method by large margins.

cs.CV↗

arXiv2023

FocalDreamer: Text-driven 3D Editing via Focal-fusion Assembly

While text-3D editing has made significant strides in leveraging score distillation sampling, emerging approaches still fall short in delivering separable, precise and consistent outcomes that are vital to content creation. In response, we introduce FocalDreamer, a framework that merges base shape with editable parts according to text prompts for fine-grained editing within desired regions. Specifically, equipped with geometry union and dual-path rendering, FocalDreamer assembles independent 3D parts into a complete object, tailored for convenient instance reuse and part-wise control. We propose geometric focal loss and style consistency regularization, which encourage focal fusion and congruent overall appearance. Furthermore, FocalDreamer generates high-fidelity geometry and PBR textures which are compatible with widely-used graphics engines. Extensive experiments have highlighted the superior editing capabilities of FocalDreamer in both quantitative and qualitative evaluations.

cs.CV↗

arXiv2023

Four years of multi-modal odometry and mapping on the rail vehicles

Precise, seamless, and efficient train localization as well as long-term railway environment monitoring is the essential property towards reliability, availability, maintainability, and safety (RAMS) engineering for railroad systems. Simultaneous localization and mapping (SLAM) is right at the core of solving the two problems concurrently. In this end, we propose a high-performance and versatile multi-modal framework in this paper, targeted for the odometry and mapping task for various rail vehicles. Our system is built atop an inertial-centric state estimator that tightly couples light detection and ranging (LiDAR), visual, optionally satellite navigation and map-based localization information with the convenience and extendibility of loosely coupled methods. The inertial sensors IMU and wheel encoder are treated as the primary sensor, which achieves the observations from subsystems to constrain the accelerometer and gyroscope biases. Compared to point-only LiDAR-inertial methods, our approach leverages more geometry information by introducing both track plane and electric power pillars into state estimation. The Visual-inertial subsystem also utilizes the environmental structure information by employing both lines and points. Besides, the method is capable of handling sensor failures by automatic reconfiguration bypassing failure modules. Our proposed method has been extensively tested in the long-during railway environments over four years, including general-speed, high-speed and metro, both passenger and freight traffic are investigated. Further, we aim to share, in an open way, the experience, problems, and successes of our group with the robotics community so that those that work in such environments can avoid these errors. In this view, we open source some of the datasets to benefit the research community.

cs.RO↗

arXiv2023

Analytical Method for Metasurface-Based Cloaking Under Arbitrary Oblique Illumination

The performance of antennas can severely deteriorate in the presence of adjacent electrically-large scatterers. In this work, we use a conducting hollow cylinder to shield the scatterer. The cylinder is shelled with single layer dielectric and electromagnetic metasurface. The scattering field analysis with respect to the surface impedance is derived. By optimizing the anisotropic impedance distribution, the scattering cross-section can be effectively reduced. The proposed method is valid for both TMz, TEz and non-TM/TE incident field. The accuracy and effectiveness of the method are verified by four cloaking scenarios in microwave regime. We demonstrate that with the surface impedance obtained by our method, a metasurface is designed with physical subwavelength structure. We also show a cloaking scenario under magnetic dipole radiation, which is closer to the case of a realistic antenna. This method can be further applied to cloaking tasks in terahertz and optical regimes.

physics.optics↗

arXiv2023

The Euclidean Space is Evil: Hyperbolic Attribute Editing for Few-shot Image Generation

Few-shot image generation is a challenging task since it aims to generate diverse new images for an unseen category with only a few images. Existing methods suffer from the trade-off between the quality and diversity of generated images. To tackle this problem, we propose Hyperbolic Attribute Editing~(HAE), a simple yet effective method. Unlike other methods that work in Euclidean space, HAE captures the hierarchy among images using data from seen categories in hyperbolic space. Given a well-trained HAE, images of unseen categories can be generated by moving the latent code of a given image toward any meaningful directions in the Poincaré disk with a fixing radius. Most importantly, the hyperbolic space allows us to control the semantic diversity of the generated images by setting different radii in the disk. Extensive experiments and visualizations demonstrate that HAE is capable of not only generating images with promising quality and diversity using limited data but achieving a highly controllable and interpretable editing process.

cs.CV↗

arXiv2023

Tackling Vision Language Tasks Through Learning Inner Monologues

Visual language tasks require AI models to comprehend and reason with both visual and textual content. Driven by the power of Large Language Models (LLMs), two prominent methods have emerged: (1) the hybrid integration between LLMs and Vision-Language Models (VLMs), where visual inputs are firstly converted into language descriptions by VLMs, serving as inputs for LLMs to generate final answer(s); (2) visual feature alignment in language space, where visual inputs are encoded as embeddings and projected to LLMs' language space via further supervised fine-tuning. The first approach provides light training costs and interpretability but is hard to be optimized in an end-to-end fashion. The second approach presents decent performance, but feature alignment usually requires large amounts of training data and lacks interpretability. To tackle this dilemma, we propose a novel approach, Inner Monologue Multi-Modal Optimization (IMMO), to solve complex vision language problems by simulating inner monologue processes, a cognitive process in which an individual engages in silent verbal communication with themselves. We enable LLMs and VLMs to interact through natural language conversation and propose to use a two-stage training process to learn how to do the inner monologue (self-asking questions and answering questions). IMMO is evaluated on two popular tasks and the results suggest by emulating the cognitive phenomenon of internal dialogue, our approach can enhance reasoning and explanation abilities, contributing to the more effective fusion of vision and language models. More importantly, instead of using predefined human-crafted monologues, IMMO learns this process within the deep learning models, promising wider applicability to many different AI problems beyond vision language tasks.

cs.CL↗

arXiv2023

3D-Aware Neural Body Fitting for Occlusion Robust 3D Human Pose Estimation

Regression-based methods for 3D human pose estimation directly predict the 3D pose parameters from a 2D image using deep networks. While achieving state-of-the-art performance on standard benchmarks, their performance degrades under occlusion. In contrast, optimization-based methods fit a parametric body model to 2D features in an iterative manner. The localized reconstruction loss can potentially make them robust to occlusion, but they suffer from the 2D-3D ambiguity. Motivated by the recent success of generative models in rigid object pose estimation, we propose 3D-aware Neural Body Fitting (3DNBF) - an approximate analysis-by-synthesis approach to 3D human pose estimation with SOTA performance and occlusion robustness. In particular, we propose a generative model of deep features based on a volumetric human representation with Gaussian ellipsoidal kernels emitting 3D pose-dependent feature vectors. The neural features are trained with contrastive learning to become 3D-aware and hence to overcome the 2D-3D ambiguity. Experiments show that 3DNBF outperforms other approaches on both occluded and standard benchmarks. Code is available at https://github.com/edz-o/3DNBF

cs.CV↗

arXiv2023

Benign Shortcut for Debiasing: Fair Visual Recognition via Intervention with Shortcut Features

Machine learning models often learn to make predictions that rely on sensitive social attributes like gender and race, which poses significant fairness risks, especially in societal applications, such as hiring, banking, and criminal justice. Existing work tackles this issue by minimizing the employed information about social attributes in models for debiasing. However, the high correlation between target task and these social attributes makes learning on the target task incompatible with debiasing. Given that model bias arises due to the learning of bias features (\emph{i.e}., gender) that help target task optimization, we explore the following research question: \emph{Can we leverage shortcut features to replace the role of bias feature in target task optimization for debiasing?} To this end, we propose \emph{Shortcut Debiasing}, to first transfer the target task's learning of bias attributes from bias features to shortcut features, and then employ causal intervention to eliminate shortcut features during inference. The key idea of \emph{Shortcut Debiasing} is to design controllable shortcut features to on one hand replace bias features in contributing to the target task during the training stage, and on the other hand be easily removed by intervention during the inference stage. This guarantees the learning of the target task does not hinder the elimination of bias features. We apply \emph{Shortcut Debiasing} to several benchmark datasets, and achieve significant improvements over the state-of-the-art debiasing methods in both accuracy and fairness.

cs.LG↗

arXiv2023

FoodSAM: Any Food Segmentation

In this paper, we explore the zero-shot capability of the Segment Anything Model (SAM) for food image segmentation. To address the lack of class-specific information in SAM-generated masks, we propose a novel framework, called FoodSAM. This innovative approach integrates the coarse semantic mask with SAM-generated masks to enhance semantic segmentation quality. Besides, we recognize that the ingredients in food can be supposed as independent individuals, which motivated us to perform instance segmentation on food images. Furthermore, FoodSAM extends its zero-shot capability to encompass panoptic segmentation by incorporating an object detector, which renders FoodSAM to effectively capture non-food object information. Drawing inspiration from the recent success of promptable segmentation, we also extend FoodSAM to promptable segmentation, supporting various prompt variants. Consequently, FoodSAM emerges as an all-encompassing solution capable of segmenting food items at multiple levels of granularity. Remarkably, this pioneering framework stands as the first-ever work to achieve instance, panoptic, and promptable segmentation on food images. Extensive experiments demonstrate the feasibility and impressing performance of FoodSAM, validating SAM's potential as a prominent and influential tool within the domain of food image segmentation. We release our code at https://github.com/jamesjg/FoodSAM.

cs.CV↗

arXiv2023

Analysis Accelerated Mirror Descent via High-resolution ODEs

Mirror descent plays a crucial role in constrained optimization and acceleration schemes, along with its corresponding low-resolution ordinary differential equations (ODEs) framework have been proposed. However, the low-resolution ODEs are unable to distinguish between Polyak's heavy-ball method and Nesterov's accelerated gradient method. This problem also arises with accelerated mirror descent. To address this issue, we derive high-resolution ODEs for accelerated mirror descent and propose a general Lyapunov function framework to analyze its convergence rate in both continuous and discrete time. Furthermore, we demonstrate that accelerated mirror descent can minimize the squared gradient norm at an inverse cubic rate.

math.OC↗

arXiv2023

PandA(Box) flies on Bluesky: maintainable and user-friendly fly scans with Mamba at HEPS

Purpose: Fly scans are indispensible in many experiments at the High Energy Photon Source (HEPS). PandABox, the main platform to implement fly scans at HEPS, needs to be integrated into Mamba, the experiment control system developed at HEPS based on Bluesky. Methods: In less than 600 lines of easily customisable and extensible backend code, provided are full control of PandABox's TCP server in native ophyd, automated configuration (also including wiring) of "PandA blocks" for constant-speed mapping experiments of various dimensions, as well as generation of scans deliberately fragmented to deal with hardware limits in numbers of exposure frames or sequencer table entries. Results: The upper-level control system for PandABox has been ported to Bluesky, enabling the combination of both components' flexibility in fly-scan applications. Based on this backend, a user-friendly Mamba frontend is developed for X-ray fluorescence (XRF) mapping experiments, which provides fully online visual feedback.

physics.ins-det↗