arXiv subjects

Qi Ma

Publications and source records attributed to Qi Ma.

29 records · Page 2Linked to original sources

arXiv2024

TechGPT-2.0: A large language model project to solve the task of knowledge graph construction

Large language models have exhibited robust performance across diverse natural language processing tasks. This report introduces TechGPT-2.0, a project designed to enhance the capabilities of large language models specifically in knowledge graph construction tasks, including named entity recognition (NER) and relationship triple extraction (RTE) tasks in NLP applications. Additionally, it serves as a LLM accessible for research within the Chinese open-source model community. We offer two 7B large language model weights and a QLoRA weight specialized for processing lengthy texts.Notably, TechGPT-2.0 is trained on Huawei's Ascend server. Inheriting all functionalities from TechGPT-1.0, it exhibits robust text processing capabilities, particularly in the domains of medicine and law. Furthermore, we introduce new capabilities to the model, enabling it to process texts in various domains such as geographical areas, transportation, organizations, literary works, biology, natural sciences, astronomical objects, and architecture. These enhancements also fortified the model's adeptness in handling hallucinations, unanswerable queries, and lengthy texts. This report provides a comprehensive and detailed introduction to the full fine-tuning process on Huawei's Ascend servers, encompassing experiences in Ascend server debugging, instruction fine-tuning data processing, and model training. Our code is available at https://github.com/neukg/TechGPT-2.0

cs.CL↗

arXiv2023

Deformable Neural Radiance Fields using RGB and Event Cameras

Modeling Neural Radiance Fields for fast-moving deformable objects from visual data alone is a challenging problem. A major issue arises due to the high deformation and low acquisition rates. To address this problem, we propose to use event cameras that offer very fast acquisition of visual change in an asynchronous manner. In this work, we develop a novel method to model the deformable neural radiance fields using RGB and event cameras. The proposed method uses the asynchronous stream of events and calibrated sparse RGB frames. In our setup, the camera pose at the individual events required to integrate them into the radiance fields remains unknown. Our method jointly optimizes these poses and the radiance field. This happens efficiently by leveraging the collection of events at once and actively sampling the events during learning. Experiments conducted on both realistically rendered graphics and real-world datasets demonstrate a significant benefit of the proposed method over the state-of-the-art and the compared baseline. This shows a promising direction for modeling deformable neural radiance fields in real-world dynamic scenes.

cs.CV↗

arXiv2023

Extremal polygonal chains with respect to the Kirchhoff index

The Kirchhoff index is defined as the sum of resistance distances between all pairs of vertices in a graph. This index is a critical parameter for measuring graph structures. In this paper, we characterize polygonal chains with the minimum Kirchhoff index, and characterize even (odd) polygonal chains with the maximum Kirchhoff index, which extends the results of \cite{45}, \cite{14} and \cite{2,13,44} to a more general case.

math.CO↗

arXiv2023

Fusing Monocular Images and Sparse IMU Signals for Real-time Human Motion Capture

Either RGB images or inertial signals have been used for the task of motion capture (mocap), but combining them together is a new and interesting topic. We believe that the combination is complementary and able to solve the inherent difficulties of using one modality input, including occlusions, extreme lighting/texture, and out-of-view for visual mocap and global drifts for inertial mocap. To this end, we propose a method that fuses monocular images and sparse IMUs for real-time human motion capture. Our method contains a dual coordinate strategy to fully explore the IMU signals with different goals in motion capture. To be specific, besides one branch transforming the IMU signals to the camera coordinate system to combine with the image information, there is another branch to learn from the IMU signals in the body root coordinate system to better estimate body poses. Furthermore, a hidden state feedback mechanism is proposed for both two branches to compensate for their own drawbacks in extreme input cases. Thus our method can easily switch between the two kinds of signals or combine them in different cases to achieve a robust mocap. %The two divided parts can help each other for better mocap results under different conditions. Quantitative and qualitative results demonstrate that by delicately designing the fusion method, our technique significantly outperforms the state-of-the-art vision, IMU, and combined methods on both global orientation and local pose estimation. Our codes are available for research at https://shaohua-pan.github.io/robustcap-page/.

cs.CV↗

arXiv2022

Extremal octagonal chains with respect to the Kirchhoff index

Let $G$ be a connected graph. The resistance distance between any two vertices of $G$ is equal to the effective resistance between them in the corresponding electrical network constructed from $G$ by replacing each edge with a unit resistor. The Kirchhoff index is defined as the sum of resistance distances between all pairs of the vertices. These indices have been computed for many interesting graphs, such as linear polyomino chain, linear/Möbius/cylinder hexagonal chain, and linear/Möbius/cylinder octagonal chain. In this paper, we characterized the maximum and minimum octagonal chains with respect to the Kirchhoff index.

math.CO↗

arXiv2022

EMFlow: Data Imputation in Latent Space via EM and Deep Flow Models

The presence of missing values within high-dimensional data is an ubiquitous problem for many applied sciences. A serious limitation of many available data mining and machine learning methods is their inability to handle partially missing values and so an integrated approach that combines imputation and model estimation is vital for down-stream analysis. A computationally fast algorithm, called EMFlow, is introduced that performs imputation in a latent space via an online version of Expectation-Maximization (EM) algorithm by using a normalizing flow (NF) model which maps the data space to a latent space. The proposed EMFlow algorithm is iterative, involving updating the parameters of online EM and NF alternatively. Extensive experimental results for high-dimensional multivariate and image datasets are presented to illustrate the superior performance of the EMFlow compared to a couple of recently available methods in terms of both predictive accuracy and speed of algorithmic convergence. We provide code for all our experiments.

cs.LG↗

arXiv2022

An Understanding-Oriented Robust Machine Reading Comprehension Model

Although existing machine reading comprehension models are making rapid progress on many datasets, they are far from robust. In this paper, we propose an understanding-oriented machine reading comprehension model to address three kinds of robustness issues, which are over sensitivity, over stability and generalization. Specifically, we first use a natural language inference module to help the model understand the accurate semantic meanings of input questions so as to address the issues of over sensitivity and over stability. Then in the machine reading comprehension module, we propose a memory-guided multi-head attention method that can further well understand the semantic meanings of input questions and passages. Third, we propose a multilanguage learning mechanism to address the issue of generalization. Finally, these modules are integrated with a multi-task learning based method. We evaluate our model on three benchmark datasets that are designed to measure models robustness, including DuReader (robust) and two SQuAD-related datasets. Extensive experiments show that our model can well address the mentioned three kinds of robustness issues. And it achieves much better results than the compared state-of-the-art models on all these datasets under different evaluation metrics, even under some extreme and unfair evaluations. The source code of our work is available at: https://github.com/neukg/RobustMRC.

cs.CL↗

arXiv2021

Predicting exoplanet mass from radius and incident flux: A Bayesian mixture model

The relationship between mass and radius (M-R relation) is the key for inferring the planetary compositions and thus valuable for the studies of formation and migration models. However, the M-R relation alone is not enough for planetary characterization due to the dependence of it on other confounding variables. This paper provides a non-trivial extension of the M-R relation by including the incident flux as an additional variable. By using Bayesian hierarchical modeling (BHM) that leverages the flexibility of finite mixture models, a probabilistic mass-radius-flux relationship (M-R-F relation) is obtained based on a sample of 319 exoplanets. We find that the flux has nonnegligible impact on the M-R relation, while such impact is strongest for hot-Jupiters. On the population level, the planets with higher level of flux tend to be denser, and high flux could trigger significant mass loss for plants with radii larger than $13R_{\oplus}$. As a result, failing to account for the flux in mass prediction would cause systematic over or under-estimation. With the recent advent of computing power, although a lot of complex statistical models can be fitted using Monte Carlo methods, it has largely remain illusive how to validate these complex models when the data are observed with large measurement errors. We present two novel methods to examine model assumptions, which can be used not only for the models we present in this paper but can also be adapted for other statistical models.

astro-ph.EP↗

arXiv2020

The normalized Laplacian and related indexes of graphs with edges blew up by cliques

In this paper, we introduce the clique-blew up graph $CL(G)$ of a given graph $G$, which is obtained from $G$ by replacing each edge of $G$ with a complete graph $K_n$. We characterize all the normalized Laplacian spectrum of the grpah $CL(G)$ in term of the given graph $G$. Based on the spectrum obtained, the formulae to calculate the multiplicative degree-Kirchhoff index, the Kemeny's constant and the number of spanning trees of $CL(G)$ are derived well. Finally, the spectrum and indexes of the clique-blew up iterative graphs are present.

math.CO↗

arXiv2020

A Non-Iterative Quantile Change Detection Method in Mixture Model with Heavy-Tailed Components

Estimating parameters of mixture model has wide applications ranging from classification problems to estimating of complex distributions. Most of the current literature on estimating the parameters of the mixture densities are based on iterative Expectation Maximization (EM) type algorithms which require the use of either taking expectations over the latent label variables or generating samples from the conditional distribution of such latent labels using the Bayes rule. Moreover, when the number of components is unknown, the problem becomes computationally more demanding due to well-known label switching issues \cite{richardson1997bayesian}. In this paper, we propose a robust and quick approach based on change-point methods to determine the number of mixture components that works for almost any location-scale families even when the components are heavy tailed (e.g., Cauchy). We present several numerical illustrations by comparing our method with some of popular methods available in the literature using simulated data and real case studies. The proposed method is shown be as much as 500 times faster than some of the competing methods and are also shown to be more accurate in estimating the mixture distributions by goodness-of-fit tests.

stat.ML↗

arXiv2019

Protocol for Asynchronous, Reliable, Secure and Efficient Consensus (PARSEC) Version 2.0

In this paper we present an open source, fully asynchronous, leaderless algorithm for reaching consensus in the presence of Byzantine faults in an asynchronous network. We prove the algorithm's correctness provided that less than a third of participating nodes are faulty. We also present a way of applying the algorithm to a network with dynamic membership, i.e. a network in which nodes can join and leave at will. The core contribution of this paper is an optimal model in the definition of an asynchronous BFT protocol, and which is resilient to 1/3 byzantine nodes. This model matches an agreement with probability one (unlike some probabilistic methods), and where a common coin is used as a source of randomization so that it respects the FLP impossibility result.

cs.CR↗