arXiv subjects

Zhen Guo

Publications and source records attributed to Zhen Guo.

At least 91 records · Page 5Linked to original sources

arXiv2024

Virtual birefringence imaging and histological staining of amyloid deposits in label-free tissue using autofluorescence microscopy and deep learning

Systemic amyloidosis is a group of diseases characterized by the deposition of misfolded proteins in various organs and tissues, leading to progressive organ dysfunction and failure. Congo red stain is the gold standard chemical stain for the visualization of amyloid deposits in tissue sections, as it forms complexes with the misfolded proteins and shows a birefringence pattern under polarized light microscopy. However, Congo red staining is tedious and costly to perform, and prone to false diagnoses due to variations in the amount of amyloid, staining quality and expert interpretation through manual examination of tissue under a polarization microscope. Here, we report the first demonstration of virtual birefringence imaging and virtual Congo red staining of label-free human tissue to show that a single trained neural network can rapidly transform autofluorescence images of label-free tissue sections into brightfield and polarized light microscopy equivalent images, matching the histochemically stained versions of the same samples. We demonstrate the efficacy of our method with blind testing and pathologist evaluations on cardiac tissue where the virtually stained images agreed well with the histochemically stained ground truth images. Our virtually stained polarization and brightfield images highlight amyloid birefringence patterns in a consistent, reproducible manner while mitigating diagnostic challenges due to variations in the quality of chemical staining and manual imaging processes as part of the clinical workflow.

physics.med-ph↗

arXiv2024

Search for Brown Dwarfs in IC 1396 with Subaru HSC: Interpreting the Impact of Environmental Factors on Sub-stellar Population

Young stellar clusters are predominantly the hub of star formation and hence, ideal to perform comprehensive studies over the least explored sub-stellar regime. Various unanswered questions like the mass distribution in brown dwarf regime and the effect of diverse cluster environment on brown dwarf formation efficiency still plague the scientific community. The nearby young cluster, IC 1396 with its feedback-driven environment, is ideal to conduct such study. In this paper we adopt a multi-wavelength approach, using deep Subaru HSC, Gaia DR3, Pan-STARRS, UKIDSS/2MASS photometry and machine learning techniques to identify the cluster members complete down to $\sim$ 0.03 M$_{\odot}$ in the central 22$^{\prime}$ area of IC 1396. We identify 458 cluster members including 62 brown dwarfs which are used to determine mass distribution in the region. We obtain a star-to-brown dwarf ratio of $\sim$ 6 for a stellar mass range 0.03 -- 1 M$_{\odot}$ in the studied cluster. The brown dwarf fraction is observed to increase across the cluster as radial distance from the central OB-stars increases. This study also compiles 15 young stellar clusters to check the variation of star-to-brown dwarf ratio relative to stellar density and UV flux ranging within 4-2500 stars pc$^{-2}$ and 0.7-7.3 G$_{0}$ respectively. The brown dwarf fraction is observed to increase with stellar density but the results about the influence of incident UV flux are inconclusive within this range. This is the deepest study of IC 1396 as of yet and it will pave the way to understand various aspects of brown dwarfs using spectroscopic observations in future.

astro-ph.SR↗

arXiv2024

Diversity Measurement and Subset Selection for Instruction Tuning Datasets

We aim to select data subsets for the fine-tuning of large language models to more effectively follow instructions. Prior work has emphasized the importance of diversity in dataset curation but relied on heuristics such as the number of tasks. In this paper, we use determinantal point processes to capture the diversity and quality of instruction tuning datasets for subset selection. We propose to measure dataset diversity with log determinant distance that is the distance between the dataset of interest and a maximally diverse reference dataset. Our experiments demonstrate that the proposed diversity measure in the normalized weight gradient space is correlated with downstream instruction-following performance. Consequently, it can be used to inform when data selection is the most helpful and to analyze dataset curation strategies. We demonstrate the utility of our approach on various instruction tuning datasets.

cs.LG↗

arXiv2024

Spectroscopic confirmation of high-amplitude eruptive YSOs and dipping giants from the VVV survey

During the pre-main-sequence (pre-MS) evolution stage of a star, significant amounts of stellar mass are accreted during episodic accretion events, such as multi-decade FUor-type outbursts. Here, we present a near-infrared spectroscopic follow-up study of 33 high-amplitude (most with $ΔK_s$ > 4 mag) variable sources discovered by the Vista Variables in the Via Lactea (VVV) survey. Based on the spectral features, 25 sources are classified as eruptive young stellar objects (YSOs), including 15 newly identified FUors, six with long-lasting but EXor-like bursts of magnetospheric accretion and four displaying outflow-dominated spectra. By examining the photometric behaviours of eruptive YSOs, we found most FUor-type outbursts have higher amplitudes ($ΔK_s$ and $ΔW2$), faster eruptive timescales and bluer infrared colours than the other outburst types. In addition, we identified seven post-main sequence variables apparently associated with deep dipping events and an eruptive star with deep AlO absorption bands resembling those seen in the V838 Mon stellar merger.

astro-ph.SR↗

arXiv2024

On the incidence of episodic accretion in Class I YSOs from VVV

Episodic accretion is one of the competing models to explain the observed luminosity spread in young stellar clusters. These short-lived high accretion events could also have a strong impact on planet formation. Observations of high-amplitude variability in young stellar objects (YSOs) due to large changes in the accretion rate provide direct observational evidence for episodic accretion. However, there are still uncertainties in the frequency of these events and if episodic accretion is universal among YSOs. To determine the frequency of outbursts in Class I YSOs, we built a large and robust sample of objects at this evolutionary stage, and searched for high-amplitude near-infrared ($ΔK_{\rm S}>2$~mag) variability in the VIRAC2 database of the Vista Variables in the Via Lactea (VVV) survey. By complementing with near-IR (2MASS and DENIS) and mid-IR (WISE/Neo-WISE) data, we find that from $\sim$ 7000 Class I YSOs, 97 objects can be classified as eruptive variable YSOs. The duration of the outbursts vary from a few months to longer than 9 years, and cover a similar range of amplitudes. Values of $ΔK_{\rm S}>5$~mag, however, are only observed in outbursts with duration longer than 9 years. When considering different effects of completeness and contamination we estimate that the incidence of episodic accretion in Class I YSOs is between 2\% and 3\%. Finally, we determine a recurrence timescale of long-term outbursts (a.k.a FUors) of $τ=1.75^{+1.12}_{-0.87}$~kyr. The latter value agrees with previous estimates and is in line with the expectations of higher frequency of FUor outbursts during younger stages of evolution.

astro-ph.SR↗

arXiv2024

Multi-wavelength detection of an ongoing FUOr-type outburst on a low-mass YSO

During the pre-main-sequence evolution, Young Stellar Objects (YSOs) assemble most of their mass during the episodic accretion process. The rarely seen FUOr-type events (FUOrs) are valuable laboratories to investigate the outbursting nature of YSOs. Here, we present multi-wavelength detection of a high-amplitude eruptive source in the young open cluster VdBH 221 with an ongoing outburst, including optical to mid-infrared time series and near-infrared spectra. The initial outburst has an exceptional amplitude of $>$6.3 mag in Gaia and 4.6 mag in $K_s$, with a peak luminosity up to 16 $L_{\odot}$ and a peak mass accretion rate of 1.4 $\times$ 10$^{-5}$ $M_\odot$ yr$^{-1}$. The optical to infrared spectral energy distribution (SED) of this object is consistent with a low-mass star (0.2$M_\odot$) with a modest extinction ($A_V < 2$ mag). A 100-d delay between optical and infrared rising stages is detected, suggesting an outside-in origin of the instability. The spectroscopic features of this object reveal a self-luminous accretion disc, very similar to FU Orionis, with a low line-of-sight extinction. Most recently, there has been a gradual increase in brightness throughout the wavelength range, possibly suggesting an enhancement of the mass accretion rate.

astro-ph.SR↗

arXiv2023

AuthentiGPT: Detecting Machine-Generated Text via Black-Box Language Models Denoising

Large language models (LLMs) have opened up enormous opportunities while simultaneously posing ethical dilemmas. One of the major concerns is their ability to create text that closely mimics human writing, which can lead to potential misuse, such as academic misconduct, disinformation, and fraud. To address this problem, we present AuthentiGPT, an efficient classifier that distinguishes between machine-generated and human-written texts. Under the assumption that human-written text resides outside the distribution of machine-generated text, AuthentiGPT leverages a black-box LLM to denoise input text with artificially added noise, and then semantically compares the denoised text with the original to determine if the content is machine-generated. With only one trainable parameter, AuthentiGPT eliminates the need for a large training dataset, watermarking the LLM's output, or computing the log-likelihood. Importantly, the detection capability of AuthentiGPT can be easily adapted to any generative language model. With a 0.918 AUROC score on a domain-specific dataset, AuthentiGPT demonstrates its effectiveness over other commercial algorithms, highlighting its potential for detecting machine-generated text in academic settings.

cs.CL↗

arXiv2023

VVV-WIT-12 and its fashionable nebula: a four year long period Young Stellar Object with a light echo?

We report the serendipitous discovery of VVV-WIT-12, an unusual variable source that seems to induce variability in its surrounding nebula. The source belongs to the rare objects that we call WITs (short for What Is This?) discovered within the VISTA Variables in the Vía Láctea (VVV) survey. VVV-WIT-12 was discovered during a pilot search for light echoes from distant Supernovae (SNe) in the Milky Way using the near-IR images of the VVV survey. This source has an extremely red spectral energy distribution, consistent with a very reddened ($A_V \sim 100$ mag) long period variable star ($P\sim1525$ days). Furthermore, it is enshrouded in a nebula that changes brightness and color with time, apparently in synch with the central source variations. The near-IR light curve and complementary follow-up spectroscopy observations are consistent with a variable Young Stellar Object (YSO) illuminating its surrounding nebula. In this case the source periodic variation along the cycles produces an unprecedented light echo in the different regions of the nebula.

astro-ph.SR↗

arXiv2023

Continuous Training and Fine-tuning for Domain-Specific Language Models in Medical Question Answering

Large language models exhibit promising general capabilities but often lack specialized knowledge for domain-specific tasks. Developing domain experts from a base model enables a range of applications without prohibitive training costs. This work demonstrates a method using continuous training and instruction fine-tuning to rapidly adapt Llama 2 base models to the Chinese medical domain. We first conduct continuous training on 1B tokens from Chinese medical references to teach relevant vocabulary and knowledge. The models are then fine-tuned on 54K examples sourced from the Chinese National Medical Licensing Examination. Experiments on Chinese medical data confirm the effectiveness of this approach, producing a model comparable to GPT-3.5-turbo while using way less computational resource. The resulting domain-specific model could be useful for various Chinese medical applications. More broadly, this provides a template for domain-specific training of large language models in areas where pre-trained models lack the required expertise, such as law, science, and engineering.

cs.CL↗

arXiv2023

Twenty-Five Years of Accretion onto the Classical T Tauri Star TW Hya

Accretion plays a central role in the physics that governs the evolution and dispersal of protoplanetary disks. The primary goal of this paper is to analyze the stability over time of the mass accretion rate onto TW Hya, the nearest accreting solar-mass young star. We measure veiling across the optical spectrum in 1169 archival high-resolution spectra of TW Hya, obtained from 1998--2022. The veiling is then converted to accretion rate using 26 flux-calibrated spectra that cover the Balmer jump. The accretion rate measured from the excess continuum has an average of $2.51\times10^{-9}$~M$_\odot$~yr$^{-1}$ and a Gaussian distribution with a FWHM of 0.22 dex. This accretion rate may be underestimated by a factor of up to 1.5 because of uncertainty in the bolometric correction and another factor of 1.7 because of excluding the fraction of accretion energy that escapes in lines, especially Ly$α$. The accretion luminosities are well correlated with He line luminosities but poorly correlated with H$α$ and H$β$ luminosity. The accretion rate is always flickering over hours but on longer timescales has been stable over 25 years. This level of variability is consistent with previous measurements for most, but not all, accreting young stars.

astro-ph.SR↗

arXiv2023

Improving Small Language Models on PubMedQA via Generative Data Augmentation

Large Language Models (LLMs) have made remarkable advancements in the field of natural language processing. However, their increasing size poses challenges in terms of computational cost. On the other hand, Small Language Models (SLMs) are known for their efficiency, but they often struggle with limited capacity and training data, especially in specific domains. In this paper, we introduce a novel method aimed at improving SLMs in the medical domain using LLM-based generative data augmentation. The objective of our approach is to develop more efficient and capable models that are specifically tailored for specialized applications. Through experiments conducted on the PubMedQA dataset, we demonstrate the effectiveness of LLMs in refining and diversifying existing question-answer pairs. This refinement process leads to improved performance in a significantly smaller model after fine-tuning. Notably, our best SLM, with under 1.6 billion parameters, outperforms the few-shot GPT-4 on the PubMedQA dataset. Our code and generated data are publicly available to facilitate further explorations.

cs.CL↗

arXiv2023

Uncertainty-Aware Reward-based Deep Reinforcement Learning for Intent Analysis of Social Media Information

Due to various and serious adverse impacts of spreading fake news, it is often known that only people with malicious intent would propagate fake news. However, it is not necessarily true based on social science studies. Distinguishing the types of fake news spreaders based on their intent is critical because it will effectively guide how to intervene to mitigate the spread of fake news with different approaches. To this end, we propose an intent classification framework that can best identify the correct intent of fake news. We will leverage deep reinforcement learning (DRL) that can optimize the structural representation of each tweet by removing noisy words from the input sequence when appending an actor to the long short-term memory (LSTM) intent classifier. Policy gradient DRL model (e.g., REINFORCE) can lead the actor to a higher delayed reward. We also devise a new uncertainty-aware immediate reward using a subjective opinion that can explicitly deal with multidimensional uncertainty for effective decision-making. Via 600K training episodes from a fake news tweets dataset with an annotated intent class, we evaluate the performance of uncertainty-aware reward in DRL. Evaluation results demonstrate that our proposed framework efficiently reduces the number of selected words to maintain a high 95\% multi-class accuracy.

cs.CL↗

arXiv2022

PPO-UE: Proximal Policy Optimization via Uncertainty-Aware Exploration

Proximal Policy Optimization (PPO) is a highly popular policy-based deep reinforcement learning (DRL) approach. However, we observe that the homogeneous exploration process in PPO could cause an unexpected stability issue in the training phase. To address this issue, we propose PPO-UE, a PPO variant equipped with self-adaptive uncertainty-aware explorations (UEs) based on a ratio uncertainty level. The proposed PPO-UE is designed to improve convergence speed and performance with an optimized ratio uncertainty level. Through extensive sensitivity analysis by varying the ratio uncertainty level, our proposed PPO-UE considerably outperforms the baseline PPO in Roboschool continuous control tasks.

cs.LG↗

arXiv2022

Noise-resilient approach for deep tomographic imaging

We propose a noise-resilient deep reconstruction algorithm for X-ray tomography. Our approach shows strong noise resilience without obtaining noisy training examples. The advantages of our framework may further enable low-photon tomographic imaging.

eess.IV↗

arXiv2022

Variational Matrix Product State Approach for Non-Hermitian System Based on a Companion Hermitian Hamiltonian

Non-Hermitian systems exhibiting topological properties are attracting growing interest. In this work, we propose an algorithm for solving the ground state of a non-Hermitian system in the matrix product state (MPS) formalism based on a companion Hermitian Hamiltonian. If the eigenvalues of the non-Hermitian system are known, the companion Hermitian Hamiltonian can be directly constructed and solved using Hermitian variational methods. When the eigenvalues are unknown, a gradient descent along with the companion Hermitian Hamiltonian yields both the ground state eigenenergy and the eigenstate. With the variational principle as a solid foundation, our algorithm ensures convergence and provides results in excellent agreement with the exact solutions of the non-Hermitian Su-Schrieffer-Heeger (nH-SSH) model as well as its interacting extension. The approach we present avoids solving any non-Hermitian matrix and overcomes numerical instabilities commonly encountered in large non-Hermitian systems.

quant-ph↗

arXiv2022

PLATO-XL: Exploring the Large-scale Pre-training of Dialogue Generation

To explore the limit of dialogue generation pre-training, we present the models of PLATO-XL with up to 11 billion parameters, trained on both Chinese and English social media conversations. To train such large models, we adopt the architecture of unified transformer with high computation and parameter efficiency. In addition, we carry out multi-party aware pre-training to better distinguish the characteristic information in social media conversations. With such designs, PLATO-XL successfully achieves superior performances as compared to other approaches in both Chinese and English chitchat. We further explore the capacity of PLATO-XL on other conversational tasks, such as knowledge grounded dialogue and task-oriented conversation. The experimental results indicate that PLATO-XL obtains state-of-the-art results across multiple conversational tasks, verifying its potential as a foundation model of conversational AI.

cs.CL↗

arXiv2022

CDConv: A Benchmark for Contradiction Detection in Chinese Conversations

Dialogue contradiction is a critical issue in open-domain dialogue systems. The contextualization nature of conversations makes dialogue contradiction detection rather challenging. In this work, we propose a benchmark for Contradiction Detection in Chinese Conversations, namely CDConv. It contains 12K multi-turn conversations annotated with three typical contradiction categories: Intra-sentence Contradiction, Role Confusion, and History Contradiction. To efficiently construct the CDConv conversations, we devise a series of methods for automatic conversation generation, which simulate common user behaviors that trigger chatbots to make contradictions. We conduct careful manual quality screening of the constructed conversations and show that state-of-the-art Chinese chatbots can be easily goaded into making contradictions. Experiments on CDConv show that properly modeling contextual information is critical for dialogue contradiction detection, but there are still unresolved challenges that require future research.

cs.CL↗

arXiv2022

A Survey on Uncertainty Reasoning and Quantification for Decision Making: Belief Theory Meets Deep Learning

An in-depth understanding of uncertainty is the first step to making effective decisions under uncertainty. Deep/machine learning (ML/DL) has been hugely leveraged to solve complex problems involved with processing high-dimensional data. However, reasoning and quantifying different types of uncertainties to achieve effective decision-making have been much less explored in ML/DL than in other Artificial Intelligence (AI) domains. In particular, belief/evidence theories have been studied in KRR since the 1960s to reason and measure uncertainties to enhance decision-making effectiveness. We found that only a few studies have leveraged the mature uncertainty research in belief/evidence theories in ML/DL to tackle complex problems under different types of uncertainty. In this survey paper, we discuss several popular belief theories and their core ideas dealing with uncertainty causes and types and quantifying them, along with the discussions of their applicability in ML/DL. In addition, we discuss three main approaches that leverage belief theories in Deep Neural Networks (DNNs), including Evidential DNNs, Fuzzy DNNs, and Rough DNNs, in terms of their uncertainty causes, types, and quantification methods along with their applicability in diverse problem domains. Based on our in-depth survey, we discuss insights, lessons learned, limitations of the current state-of-the-art bridging belief theories and ML/DL, and finally, future research directions.

cs.AI↗