Searcharxiv⌕ Search

arXiv subjects

Yan Xu

Publications and source records attributed to Yan Xu.

At least 145 records · Page 8Linked to original sources
arXiv2023

Multi-View Person Matching and 3D Pose Estimation with Arbitrary Uncalibrated Camera Networks

Cross-view person matching and 3D human pose estimation in multi-camera networks are particularly difficult when the cameras are extrinsically uncalibrated. Existing efforts generally require large amounts of 3D data for training neural networks or known camera poses for geometric constraints to solve the problem. However, camera poses and 3D data annotation are usually expensive and not always available. We present a method, PME, that solves the two tasks without requiring either information. Our idea is to address cross-view person matching as a clustering problem using each person as a cluster center, then obtain correspondences from person matches, and estimate 3D human poses through multi-view triangulation and bundle adjustment. We solve the clustering problem by introducing a "size constraint" using the number of cameras and a "source constraint" using the fact that two people from the same camera view should not match, to narrow the solution space to a small feasible region. The 2D human poses used in clustering are obtained through a pre-trained 2D pose detector, so our method does not require expensive 3D training data for each new scene. We extensively evaluate our method on three open datasets and two indoor and outdoor datasets collected using arbitrarily set cameras. Our method outperforms other methods by a large margin on cross-view person matching, reaches SOTA performance on 3D human pose estimation without using either camera poses or 3D training data, and shows good generalization ability across five datasets of various environment settings.

cs.CV↗
arXiv2023

Persistent hypergraph homology and its applications

Persistent homology theory is a relatively new but powerful method in data analysis. Using simplicial complexes, classical persistent homology is able to reveal high dimensional geometric structures of datasets, and represent them as persistent barcodes. However, many datasets contain complex systems of multi-way interactions, making these datasets more naturally and faithfully modeled by hypergraphs. In this article, we investigate the persistent hypergraph model, an important generalization of the classical persistent homology on simplicial complexes. We introduce a new homology, $\hat{H}$, on hypergraphs and an efficient algorithm to compute both persistent barcodes and $\hat{H}$ barcodes. As example, our theory is demonstrated by analyzing face-to-face interactions of different populations. The datasets that we select consist of baboons in primate center, people from rural Malawi, scientific conference, workplace and high school.

math.AT↗
arXiv2023

A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity

This paper proposes a framework for quantitatively evaluating interactive LLMs such as ChatGPT using publicly available data sets. We carry out an extensive technical evaluation of ChatGPT using 23 data sets covering 8 different common NLP application tasks. We evaluate the multitask, multilingual and multi-modal aspects of ChatGPT based on these data sets and a newly designed multimodal dataset. We find that ChatGPT outperforms LLMs with zero-shot learning on most tasks and even outperforms fine-tuned models on some tasks. We find that it is better at understanding non-Latin script languages than generating them. It is able to generate multimodal content from textual prompts, via an intermediate code generation step. Moreover, we find that ChatGPT is 63.41% accurate on average in 10 different reasoning categories under logical reasoning, non-textual reasoning, and commonsense reasoning, hence making it an unreliable reasoner. It is, for example, better at deductive than inductive reasoning. ChatGPT suffers from hallucination problems like other LLMs and it generates more extrinsic hallucinations from its parametric memory as it does not have access to an external knowledge base. Finally, the interactive feature of ChatGPT enables human collaboration with the underlying LLM to improve its performance, i.e, 8% ROUGE-1 on summarization and 2% ChrF++ on machine translation, in a multi-turn "prompt engineering" fashion. We also release codebase for evaluation set extraction.

cs.CL↗
arXiv2023

A high-order local discontinuous Galerkin method for the $p$-Laplace equation

We study the high-order local discontinuous Galerkin (LDG) method for the $p$-Laplace equation. We reformulate our spatial discretization as an equivalent convex minimization problem and use a preconditioned gradient descent method as the nonlinear solver. For the first time, a weighted preconditioner that provides $hk$-independent convergence is applied in the LDG setting. For polynomial order $k \geqslant 1$, we rigorously establish the solvability of our scheme and provide a priori error estimates in a mesh-dependent energy norm. Our error estimates are under a different and non-equivalent distance from existing LDG results. For arbitrarily high-order polynomials under the assumption that the exact solution has enough regularity, the error estimates demonstrate the potential for high-order accuracy. Our numerical results exhibit the desired convergence speed facilitated by the preconditioner, and we observe best convergence rates in gradient variables in alignment with linear LDG, and optimal rates in the primal variable when $1 < p \leqslant 2$.

math.NA↗
arXiv2023

Contrastive Learning for Inference in Dialogue

Inference, especially those derived from inductive processes, is a crucial component in our conversation to complement the information implicitly or explicitly conveyed by a speaker. While recent large language models show remarkable advances in inference tasks, their performance in inductive reasoning, where not all information is present in the context, is far behind deductive reasoning. In this paper, we analyze the behavior of the models based on the task difficulty defined by the semantic information gap -- which distinguishes inductive and deductive reasoning (Johnson-Laird, 1988, 1993). Our analysis reveals that the disparity in information between dialogue contexts and desired inferences poses a significant challenge to the inductive inference process. To mitigate this information gap, we investigate a contrastive learning approach by feeding negative samples. Our experiments suggest negative samples help models understand what is wrong and improve their inference generations.

cs.CL↗
arXiv2023

Deep learning based on Transformer architecture for power system short-term voltage stability assessment with class imbalance

Most existing data-driven power system short-term voltage stability assessment (STVSA) approaches presume class-balanced input data. However, in practical applications, the occurrence of short-term voltage instability following a disturbance is minimal, leading to a significant class imbalance problem and a consequent decline in classifier performance. This work proposes a Transformer-based STVSA method to address this challenge. By utilizing the basic Transformer architecture, a stability assessment Transformer (StaaT) is developed {as a classification model to reflect the correlation between the operational states of the system and the resulting stability outcomes}. To combat the negative impact of imbalanced datasets, this work employs a conditional Wasserstein generative adversarial network with gradient penalty (CWGAN-GP) for synthetic data generation, aiding in the creation of a balanced, representative training set for the classifier. Semi-supervised clustering learning is implemented to enhance clustering quality, addressing the lack of a unified quantitative criterion for short-term voltage stability. {Numerical tests on the IEEE 39-bus test system extensively demonstrate that the proposed method exhibits robust performance under class imbalances up to 100:1 and noisy environments, and maintains consistent effectiveness even with an increased penetration of renewable energy}. Comparative results reveal that the CWGAN-GP generates more balanced datasets than traditional oversampling methods and that the StaaT outperforms other deep learning algorithms. This study presents a compelling solution for real-world STVSA applications that often face class imbalance and data noise challenges.

eess.SY↗
arXiv2023

Towards Mitigating Hallucination in Large Language Models via Self-Reflection

Large language models (LLMs) have shown promise for generative and knowledge-intensive tasks including question-answering (QA) tasks. However, the practical deployment still faces challenges, notably the issue of "hallucination", where models generate plausible-sounding but unfaithful or nonsensical information. This issue becomes particularly critical in the medical domain due to the uncommon professional concepts and potential social risks involved. This paper analyses the phenomenon of hallucination in medical generative QA systems using widely adopted LLMs and datasets. Our investigation centers on the identification and comprehension of common problematic answers, with a specific emphasis on hallucination. To tackle this challenge, we present an interactive self-reflection methodology that incorporates knowledge acquisition and answer generation. Through this feedback process, our approach steadily enhances the factuality, consistency, and entailment of the generated answers. Consequently, we harness the interactivity and multitasking ability of LLMs and produce progressively more precise and accurate answers. Experimental results on both automatic and human evaluation demonstrate the superiority of our approach in hallucination reduction compared to baselines.

cs.CL↗
arXiv2023

PICK: Polished & Informed Candidate Scoring for Knowledge-Grounded Dialogue Systems

Grounding dialogue response generation on external knowledge is proposed to produce informative and engaging responses. However, current knowledge-grounded dialogue (KGD) systems often fail to align the generated responses with human-preferred qualities due to several issues like hallucination and the lack of coherence. Upon analyzing multiple language model generations, we observe the presence of alternative generated responses within a single decoding process. These alternative responses are more faithful and exhibit a comparable or higher level of relevance to prior conversational turns compared to the optimal responses prioritized by the decoding processes. To address these challenges and driven by these observations, we propose Polished \& Informed Candidate Scoring (PICK), a generation re-scoring framework that empowers models to generate faithful and relevant responses without requiring additional labeled data or model tuning. Through comprehensive automatic and human evaluations, we demonstrate the effectiveness of PICK in generating responses that are more faithful while keeping them relevant to the dialogue history. Furthermore, PICK consistently improves the system's performance with both oracle and retrieved knowledge in all decoding strategies. We provide the detailed implementation in https://github.com/bryanwilie/pick .

cs.CL↗
arXiv2023

XR Input Error Mediation for Hand-Based Input: Task and Context Influences a User's Preference

Many XR devices use bare-hand gestures to reduce the need for handheld controllers. Such gestures, however, lead to false positive and false negative recognition errors, which detract from the user experience. While mediation techniques enable users to overcome recognition errors by clarifying their intentions via UI elements, little research has explored how mediation techniques should be designed in XR and how a user's task and context may impact their design preferences. This research presents empirical studies about the impact of user perceived error costs on users' preferences for three mediation technique designs, under different simulated scenarios that were inspired by real-life tasks. Based on a large-scale crowd-sourced survey and an immersive VR-based user study, our results suggest that the varying contexts within each task type can impact users' perceived error costs, leading to different preferred mediation techniques. We further discuss the study implications of these results on future XR interaction design.

cs.HC↗
arXiv2023

Nucleus-aware Self-supervised Pretraining Using Unpaired Image-to-image Translation for Histopathology Images

Self-supervised pretraining attempts to enhance model performance by obtaining effective features from unlabeled data, and has demonstrated its effectiveness in the field of histopathology images. Despite its success, few works concentrate on the extraction of nucleus-level information, which is essential for pathologic analysis. In this work, we propose a novel nucleus-aware self-supervised pretraining framework for histopathology images. The framework aims to capture the nuclear morphology and distribution information through unpaired image-to-image translation between histopathology images and pseudo mask images. The generation process is modulated by both conditional and stochastic style representations, ensuring the reality and diversity of the generated histopathology images for pretraining. Further, an instance segmentation guided strategy is employed to capture instance-level information. The experiments on 7 datasets show that the proposed pretraining method outperforms supervised ones on Kather classification, multiple instance learning, and 5 dense-prediction tasks with the transfer learning protocol, and yields superior results than other self-supervised approaches on 8 semi-supervised tasks. Our project is publicly available at https://github.com/zhiyuns/UNITPathSSL.

cs.CV↗
arXiv2023

NICE: CVPR 2023 Challenge on Zero-shot Image Captioning

In this report, we introduce NICE (New frontiers for zero-shot Image Captioning Evaluation) project and share the results and outcomes of 2023 challenge. This project is designed to challenge the computer vision community to develop robust image captioning models that advance the state-of-the-art both in terms of accuracy and fairness. Through the challenge, the image captioning models were tested using a new evaluation dataset that includes a large variety of visual concepts from many domains. There was no specific training data provided for the challenge, and therefore the challenge entries were required to adapt to new types of image descriptions that had not been seen during training. This report includes information on the newly proposed NICE dataset, evaluation methods, challenge results, and technical details of top-ranking entries. We expect that the outcomes of the challenge will contribute to the improvement of AI models on various vision-language tasks.

cs.CV↗
arXiv2023

Extreme Red-wing Enhancements of UV Lines During the 2022 March 30 X1.3 Solar Flare

Here we present the study of a compact emission source during an X1.3 flare on 2022-March-30. Within a $\sim41$~s period (17:34:48 UT to 17:35:29 UT), IRIS observations show spectral lines of Mg II, C II and Si IV with extremely broadened, asymmetric red-wings. This source of interest (SOI) is compact, $\sim$ 1\arcsec.6, and is located in the wake of a passing ribbon. Two methods were applied to measure the Doppler velocities associated with these red wings: spectral moments and multi-Gaussian fits. The spectral moments method considers the averaged shift of the lines, which are 85 km s$^{-1}$, 125 km s$^{-1}$ and 115 km s$^{-1}$ for the Mg II, C II and Si IV lines respectively. The red-most Gaussian fit suggests a Doppler velocity up to $\sim$160 km s$^{-1}$ in all of the three lines. Downward mass motions with such high speeds are very atypical, with most chromospheric downflows in flares on the order 10-100 km s$^{-1}$. Furthermore, EUV emission is strong within flaring loops connecting two flare ribbons located mainly to the east of the central flare region. The EUV loops that connect the SOI and its counterpart source in the opposite field are much less brightened, indicating that the density and/or temperature is comparatively low. These observations suggest a very fast downflowing plasma in transition region and upper chromosphere, that decelerates rapidly since there is no equivalently strong shift of the O I chromospheric lines. This unusual observation presents a challenge that models of the solar atmosphere's response to flares must be able to explain.

astro-ph.SR↗
arXiv2023

One-Shot Strategically Deconflicted Route and Operational Volume Generation for Urban Air Mobility Operations

In the UAM space, strategic deconfliction provides an all-essential layer to airspace automation by providing safe, pre-emptive deconfliction or assignment of airspace resources to airspace users pre-flight. Strategic deconfliction approaches provide an elegant solution to pre-flight deconfliction operations. This overall creates safer and more efficient airspace and reduces the workload on controllers. In this research, we propose a method that constructs routes between start and end nodes in airspace, assigns a contract of operational volumes (OVs) and ensures that these OVs are sufficiently deconflicted against static no-fly zones and OVs of other airspace users. Our approach uses the A* optimal cost path algorithm to generate the shortest routes between the origin and destination. We present a method for generating OVs based on the distribution of aircraft positions from simulated flights; volumes are constructed such that this distribution is conservatively described.

cs.RO↗
arXiv2023

Physics-Informed Deep Learning to Reduce the Bias in Joint Prediction of Nitrogen Oxides

Atmospheric nitrogen oxides (NOx) primarily from fuel combustion have recognized acute and chronic health and environmental effects. Machine learning (ML) methods have significantly enhanced our capacity to predict NOx concentrations at ground-level with high spatiotemporal resolution but may suffer from high estimation bias since they lack physical and chemical knowledge about air pollution dynamics. Chemical transport models (CTMs) leverage this knowledge; however, accurate predictions of ground-level concentrations typically necessitate extensive post-calibration. Here, we present a physics-informed deep learning framework that encodes advection-diffusion mechanisms and fluid dynamics constraints to jointly predict NO2 and NOx and reduce ML model bias by 21-42%. Our approach captures fine-scale transport of NO2 and NOx, generates robust spatial extrapolation, and provides explicit uncertainty estimation. The framework fuses knowledge-driven physicochemical principles of CTMs with the predictive power of ML for air quality exposure, health, and policy applications. Our approach offers significant improvements over purely data-driven ML methods and has unprecedented bias reduction in joint NO2 and NOx prediction.

cs.LG↗
arXiv2023

Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference

The capability to generate responses with diversity and faithfulness using factual knowledge is paramount for creating a human-like, trustworthy dialogue system. Common strategies either adopt a two-step paradigm, which optimizes knowledge selection and response generation separately, and may overlook the inherent correlation between these two tasks, or leverage conditional variational method to jointly optimize knowledge selection and response generation by employing an inference network. In this paper, we present an end-to-end learning framework, termed Sequential Posterior Inference (SPI), capable of selecting knowledge and generating dialogues by approximately sampling from the posterior distribution. Unlike other methods, SPI does not require the inference network or assume a simple geometry of the posterior distribution. This straightforward and intuitive inference procedure of SPI directly queries the response generation model, allowing for accurate knowledge selection and generation of faithful responses. In addition to modeling contributions, our experimental results on two common dialogue datasets (Wizard of Wikipedia and Holl-E) demonstrate that SPI outperforms previous strong baselines according to both automatic and human evaluation metrics.

cs.CL↗
arXiv2023

Elastically-Constrained Meta-Learner for Federated Learning

Federated learning is an approach to collaboratively training machine learning models for multiple parties that prohibit data sharing. One of the challenges in federated learning is non-IID data between clients, as a single model can not fit the data distribution for all clients. Meta-learning, such as Per-FedAvg, is introduced to cope with the challenge. Meta-learning learns shared initial parameters for all clients. Each client employs gradient descent to adapt the initialization to local data distributions quickly to realize model personalization. However, due to non-convex loss function and randomness of sampling update, meta-learning approaches have unstable goals in local adaptation for the same client. This fluctuation in different adaptation directions hinders the convergence in meta-learning. To overcome this challenge, we use the historical local adapted model to restrict the direction of the inner loop and propose an elastic-constrained method. As a result, the current round inner loop keeps historical goals and adapts to better solutions. Experiments show our method boosts meta-learning convergence and improves personalization without additional calculation and communication. Our method achieved SOTA on all metrics in three public datasets.

cs.LG↗