arXiv subjects

Xinyu Dai

Publications and source records attributed to Xinyu Dai.

At least 91 records · Page 5Linked to original sources

arXiv2022

Analyzing the Intensity of Complaints on Social Media

Complaining is a speech act that expresses a negative inconsistency between reality and human expectations. While prior studies mostly focus on identifying the existence or the type of complaints, in this work, we present the first study in computational linguistics of measuring the intensity of complaints from text. Analyzing complaints from such perspective is particularly useful, as complaints of certain degrees may cause severe consequences for companies or organizations. We create the first Chinese dataset containing 3,103 posts about complaints from Weibo, a popular Chinese social media platform. These posts are then annotated with complaints intensity scores using Best-Worst Scaling (BWS) method. We show that complaints intensity can be accurately estimated by computational models with the best mean square error achieving 0.11. Furthermore, we conduct a comprehensive linguistic analysis around complaints, including the connections between complaints and sentiment, and a cross-lingual comparison for complaints expressions used by Chinese and English speakers. We finally show that our complaints intensity scores can be incorporated for better estimating the popularity of posts on social media.

cs.CL↗

arXiv2022

$\textit{latent}$-GLAT: Glancing at Latent Variables for Parallel Text Generation

Recently, parallel text generation has received widespread attention due to its success in generation efficiency. Although many advanced techniques are proposed to improve its generation quality, they still need the help of an autoregressive model for training to overcome the one-to-many multi-modal phenomenon in the dataset, limiting their applications. In this paper, we propose $\textit{latent}$-GLAT, which employs the discrete latent variables to capture word categorical information and invoke an advanced curriculum learning technique, alleviating the multi-modality problem. Experiment results show that our method outperforms strong baselines without the help of an autoregressive model, which further broadens the application scenarios of the parallel decoding paradigm.

cs.CL↗

arXiv2021

Conditional Neural Process for non-parametric modeling of AGN light curve

The consequences of complex disturbed environments in the vicinity of a supermassive black hole are not well represented by standard statistical models of optical variability in active galactic nuclei (AGN). Thus, developing new methodologies for investigating and modeling AGN light curves is crucial. Conditional Neural Processes (CNPs) are nonlinear function models that forecast stochastic time-series based on a finite amount of known data without the use of any additional parameters or prior knowledge (kernels). We provide a CNP algorithm that is specifically designed for simulating AGN light curves. It was trained using data from the All-Sky Automated Survey for Supernovae, which included 153 AGN. We present CNP modeling performance for a subsample of five AGNs with distinctive difficult-to-model properties. The performance of CNP in predicting temporal flux fluctuation was assessed using a minimizing loss function, and the results demonstrated the algorithm's usefulness. Our preliminary parallelization experiments show that CNP can efficiently handle large amounts of data. These results imply that CNP can be more effective than standard tools in modeling large volumes of AGN data (as anticipated from time-domain surveys such as the Vera C. Rubin Observatory's Legacy Survey of Space and Time).

astro-ph.GA↗

arXiv2021

Optical Confirmation of X-ray selected Galaxy clusters from the Swift AGN and Cluster survey with MDM and Pan-STARRS Data (Paper III)

To understand structure formation in the universe and impose stronger constraints on the cluster mass function and cosmological models, it is important to have large galaxy cluster catalogs. The Swift AGN and Cluster Survey is a serendipitous X-ray survey aimed at building a large statistically selected X-ray cluster catalog with 442 cluster candidates in its first release. Our initial SDSS follow-up study confirmed $50\%$ of clusters in the SDSS footprint as z $<$ 0.5 clusters. Here, we present further optical follow-up analysis of 248 (out of 442) cluster candidates from the Swift cluster catalog using multi-band imaging from the MDM $2.4m$ telescope and the Pan-STARRS survey. We report the optical confirmation of 55 clusters with $> 3σ$ galaxy overdensities and detectable red sequences in the color-magnitude space. The majority of these confirmed clusters have redshifts z $<$ 0.6. The remaining candidates are potentially higher redshift clusters that are excellent targets for infrared observations. We report the X-ray luminosity and the optical richness for these confirmed clusters. We also discuss the distinction between X-ray and optical observables for the detected and non-detected cluster candidates.

astro-ph.GA↗

arXiv2021

Energy-based Unknown Intent Detection with Data Manipulation

Unknown intent detection aims to identify the out-of-distribution (OOD) utterance whose intent has never appeared in the training set. In this paper, we propose using energy scores for this task as the energy score is theoretically aligned with the density of the input and can be derived from any classifier. However, high-quality OOD utterances are required during the training stage in order to shape the energy gap between OOD and in-distribution (IND), and these utterances are difficult to collect in practice. To tackle this problem, we propose a data manipulation framework to Generate high-quality OOD utterances with importance weighTs (GOT). Experimental results show that the energy-based detector fine-tuned by GOT can achieve state-of-the-art results on two benchmark datasets.

cs.CL↗

arXiv2021

UniDrop: A Simple yet Effective Technique to Improve Transformer without Extra Cost

Transformer architecture achieves great success in abundant natural language processing tasks. The over-parameterization of the Transformer model has motivated plenty of works to alleviate its overfitting for superior performances. With some explorations, we find simple techniques such as dropout, can greatly boost model performance with a careful design. Therefore, in this paper, we integrate different dropout techniques into the training of Transformer models. Specifically, we propose an approach named UniDrop to unites three different dropout techniques from fine-grain to coarse-grain, i.e., feature dropout, structure dropout, and data dropout. Theoretically, we demonstrate that these three dropouts play different roles from regularization perspectives. Empirically, we conduct experiments on both neural machine translation and text classification benchmark datasets. Extensive results indicate that Transformer with UniDrop can achieve around 1.5 BLEU improvement on IWSLT14 translation tasks, and better accuracy for the classification even using strong pre-trained RoBERTa as backbone.

cs.CL↗

arXiv2021

Non-Autoregressive Translation by Learning Target Categorical Codes

Non-autoregressive Transformer is a promising text generation model. However, current non-autoregressive models still fall behind their autoregressive counterparts in translation quality. We attribute this accuracy gap to the lack of dependency modeling among decoder inputs. In this paper, we propose CNAT, which learns implicitly categorical codes as latent variables into the non-autoregressive decoding. The interaction among these categorical codes remedies the missing dependencies and improves the model capacity. Experiment results show that our model achieves comparable or better performance in machine translation tasks, compared with several strong baselines.

cs.CL↗

arXiv2021

Dual Side Deep Context-aware Modulation for Social Recommendation

Social recommendation is effective in improving the recommendation performance by leveraging social relations from online social networking platforms. Social relations among users provide friends' information for modeling users' interest in candidate items and help items expose to potential consumers (i.e., item attraction). However, there are two issues haven't been well-studied: Firstly, for the user interests, existing methods typically aggregate friends' information contextualized on the candidate item only, and this shallow context-aware aggregation makes them suffer from the limited friends' information. Secondly, for the item attraction, if the item's past consumers are the friends of or have a similar consumption habit to the targeted user, the item may be more attractive to the targeted user, but most existing methods neglect the relation enhanced context-aware item attraction. To address the above issues, we proposed DICER (Dual Side Deep Context-aware Modulation for SocialRecommendation). Specifically, we first proposed a novel graph neural network to model the social relation and collaborative relation, and on top of high-order relations, a dual side deep context-aware modulation is introduced to capture the friends' information and item attraction. Empirical results on two real-world datasets show the effectiveness of the proposed model and further experiments are conducted to help understand how the dual context-aware modulation works.

cs.IR↗

arXiv2021

The Changing Look Blazar B2 1420+32

Blazars are active galactic nuclei with their relativistic jets pointing toward the observer, with two major sub-classes, the flat spectrum radio quasars and BL Lac objects. We present multi-wavelength photometric and spectroscopic monitoring observations of the blazar, B2 1420+32, focusing on its outbursts in 2018-2020. Multi-epoch spectra show that the blazar exhibited large scale spectral variability in both its continuum and line emission, accompanied by dramatic gamma-ray and optical variability by factors of up to 40 and 15, respectively, on week to month timescales. Over the last decade, the gamma-ray and optical fluxes increased by factors of 1500 and 100, respectively. B2 1420+32 was an FSRQ with broad emission lines in 1995. Following a series of flares starting in 2018, it transitioned between BL Lac and FSRQ states multiple times, with the emergence of a strong Fe pseudo continuum. Two spectra also contain components that can be modeled as single-temperature black bodies of 12,000 and 5,200 K. Such a collection of "changing look" features has never been observed previously in a blazar. We measure gamma-ray-optical and the inter-band optical lags implying emission region separations of less than 800 and 130 gravitational radii respectively. Since most emission line flux variations, except the Fe continuum, are within a factor of 2-3, the transitions between FSRQ and BL Lac classifications are mainly caused by the continuum variability. The large Fe continuum flux increase suggests the occurrence of dust sublimation releasing more Fe ions in the central engine and an energy transfer from the relativistic jet to sub-relativistic emission components.

astro-ph.HE↗

arXiv2020

Suzaku Measurements of Hot Halo Emission at Outskirts for Two Poor Galaxy Groups: NGC 3402 and NGC 5129

We present Suzaku off-center observations of two poor galaxy groups, NGC 3402 and NGC 5129, with temperatures below 1 keV. Through spectral decomposition, we measured their surface brightnesses and temperatures out to 530 and 1430 times the critical density of the universe for NGC 3402 and NGC 5129, respectively. These quantities are consistent with extrapolations from existing inner measurements of the two groups. With the refined bolometric X-ray luminosities, both groups prefer $L_X$-$T$ relations without a break in the group regime. Furthermore, we have determined the electron number densities and hydrostatic masses at these radii. We found that the surface brightness and electron number density profiles require two $β$ model components, as well as the indication that a third $β$ model may be needed for NGC 3402. Adding the gas mass measured from the X-ray data and stellar mass from group galaxy members, we computed baryon fractions of $f_b$ = 0.0693 $\pm$ 0.0068 and $f_b$ = 0.095 $\pm$ 0.014 for NGC 3402 and NGC 5129, respectively. Combining other poor groups with well-measured X-ray emission to the outskirts, we found an average baryon fraction extrapolated to $r_{500}$ of $\overline{f_{b,500}}$ = 0.0912 $\pm$ 0.0050 for X-ray-bright groups with temperatures between 0.8 and 1.3 keV, extending existing constraints to lower-mass systems and indicating that significant baryon losses exist below approximately $r_{500}$.

astro-ph.GA↗

arXiv2020

Grid Tagging Scheme for Aspect-oriented Fine-grained Opinion Extraction

Aspect-oriented Fine-grained Opinion Extraction (AFOE) aims at extracting aspect terms and opinion terms from review in the form of opinion pairs or additionally extracting sentiment polarity of aspect term to form opinion triplet. Because of containing several opinion factors, the complete AFOE task is usually divided into multiple subtasks and achieved in the pipeline. However, pipeline approaches easily suffer from error propagation and inconvenience in real-world scenarios. To this end, we propose a novel tagging scheme, Grid Tagging Scheme (GTS), to address the AFOE task in an end-to-end fashion only with one unified grid tagging task. Additionally, we design an effective inference strategy on GTS to exploit mutual indication between different opinion factors for more accurate extractions. To validate the feasibility and compatibility of GTS, we implement three different GTS models respectively based on CNN, BiLSTM, and BERT, and conduct experiments on the aspect-oriented opinion pair extraction and opinion triplet extraction datasets. Extensive experimental results indicate that GTS models outperform strong baselines significantly and achieve state-of-the-art performance.

cs.CL↗

arXiv2020

Opinion Transmission Network for Jointly Improving Aspect-oriented Opinion Words Extraction and Sentiment Classification

Aspect-level sentiment classification (ALSC) and aspect oriented opinion words extraction (AOWE) are two highly relevant aspect-based sentiment analysis (ABSA) subtasks. They respectively aim to detect the sentiment polarity and extract the corresponding opinion words toward a given aspect in a sentence. Previous works separate them and focus on one of them by training neural models on small-scale labeled data, while neglecting the connections between them. In this paper, we propose a novel joint model, Opinion Transmission Network (OTN), to exploit the potential bridge between ALSC and AOWE to achieve the goal of facilitating them simultaneously. Specifically, we design two tailor-made opinion transmission mechanisms to control opinion clues flow bidirectionally, respectively from ALSC to AOWE and AOWE to ALSC. Experiment results on two benchmark datasets show that our joint model outperforms strong baselines on the two tasks. Further analysis also validates the effectiveness of opinion transmission mechanisms.

cs.CL↗

arXiv2020

Transformer-based Multi-Aspect Modeling for Multi-Aspect Multi-Sentiment Analysis

Aspect-based sentiment analysis (ABSA) aims at analyzing the sentiment of a given aspect in a sentence. Recently, neural network-based methods have achieved promising results in existing ABSA datasets. However, these datasets tend to degenerate to sentence-level sentiment analysis because most sentences contain only one aspect or multiple aspects with the same sentiment polarity. To facilitate the research of ABSA, NLPCC 2020 Shared Task 2 releases a new large-scale Multi-Aspect Multi-Sentiment (MAMS) dataset. In the MAMS dataset, each sentence contains at least two different aspects with different sentiment polarities, which makes ABSA more complex and challenging. To address the challenging dataset, we re-formalize ABSA as a problem of multi-aspect sentiment analysis, and propose a novel Transformer-based Multi-aspect Modeling scheme (TMM), which can capture potential relations between multiple aspects and simultaneously detect the sentiment of all aspects in a sentence. Experiment results on the MAMS dataset show that our method achieves noticeable improvements compared with strong baselines such as BERT and RoBERTa, and finally ranks the 2nd in NLPCC 2020 Shared Task 2 Evaluation.

cs.CL↗

arXiv2020

Attention Transfer Network for Aspect-level Sentiment Classification

Aspect-level sentiment classification (ASC) aims to detect the sentiment polarity of a given opinion target in a sentence. In neural network-based methods for ASC, most works employ the attention mechanism to capture the corresponding sentiment words of the opinion target, then aggregate them as evidence to infer the sentiment of the target. However, aspect-level datasets are all relatively small-scale due to the complexity of annotation. Data scarcity causes the attention mechanism sometimes to fail to focus on the corresponding sentiment words of the target, which finally weakens the performance of neural models. To address the issue, we propose a novel Attention Transfer Network (ATN) in this paper, which can successfully exploit attention knowledge from resource-rich document-level sentiment classification datasets to improve the attention capability of the aspect-level sentiment classification task. In the ATN model, we design two different methods to transfer attention knowledge and conduct experiments on two ASC benchmark datasets. Extensive experimental results show that our methods consistently outperform state-of-the-art works. Further analysis also validates the effectiveness of ATN.

cs.CL↗

arXiv2020

Prompt Agnostic Essay Scorer: A Domain Generalization Approach to Cross-prompt Automated Essay Scoring

Cross-prompt automated essay scoring (AES) requires the system to use non target-prompt essays to award scores to a target-prompt essay. Since obtaining a large quantity of pre-graded essays to a particular prompt is often difficult and unrealistic, the task of cross-prompt AES is vital for the development of real-world AES systems, yet it remains an under-explored area of research. Models designed for prompt-specific AES rely heavily on prompt-specific knowledge and perform poorly in the cross-prompt setting, whereas current approaches to cross-prompt AES either require a certain quantity of labelled target-prompt essays or require a large quantity of unlabelled target-prompt essays to perform transfer learning in a multi-step manner. To address these issues, we introduce Prompt Agnostic Essay Scorer (PAES) for cross-prompt AES. Our method requires no access to labelled or unlabelled target-prompt data during training and is a single-stage approach. PAES is easy to apply in practice and achieves state-of-the-art performance on the Automated Student Assessment Prize (ASAP) dataset.

cs.CL↗

arXiv2020

Adding A Filter Based on The Discriminator to Improve Unconditional Text Generation

The autoregressive language model (ALM) trained with maximum likelihood estimation (MLE) is widely used in unconditional text generation. Due to exposure bias, the generated texts still suffer from low quality and diversity. This presents statistically as a discrepancy between the real text and generated text. Some research shows a discriminator can detect this discrepancy. Because the discriminator can encode more information than the generator, discriminator has the potentiality to improve generator. To alleviate the exposure bias, generative adversarial networks (GAN) use the discriminator to update the generator's parameters directly, but they fail by being evaluated precisely. A critical reason for the failure is the difference between the discriminator input and the ALM input. We propose a novel mechanism by adding a filter which has the same input as the discriminator. First, discriminator detects the discrepancy signals and passes to filter directly (or by learning). Then, we use the filter to reject some generated samples with a sampling-based method. Thus, the original generative distribution is revised to reduce the discrepancy. Two ALMs, RNN-based and Transformer-based, are experimented. Evaluated precisely by three metrics, our mechanism consistently outperforms the ALMs and all kinds of GANs across two benchmark data sets.

cs.CV↗

arXiv2020

A Reinforced Generation of Adversarial Examples for Neural Machine Translation

Neural machine translation systems tend to fail on less decent inputs despite its significant efficacy, which may significantly harm the credibility of this systems-fathoming how and when neural-based systems fail in such cases is critical for industrial maintenance. Instead of collecting and analyzing bad cases using limited handcrafted error features, here we investigate this issue by generating adversarial examples via a new paradigm based on reinforcement learning. Our paradigm could expose pitfalls for a given performance metric, e.g., BLEU, and could target any given neural machine translation architecture. We conduct experiments of adversarial attacks on two mainstream neural machine translation architectures, RNN-search, and Transformer. The results show that our method efficiently produces stable attacks with meaning-preserving adversarial examples. We also present a qualitative and quantitative analysis for the preference pattern of the attack, demonstrating its capability of pitfall exposure.

cs.CL↗

arXiv2020

Identifying the Lens Galaxy B1152+199 as a Ghostly Damped Lyman Alpha System by the Cosmic Origins Spectrograph

Strong quasar-galaxy lensing provides a powerful tool to probe the inter-stellar medium (ISM) of the lens galaxy using radiation from the background quasar. Using the Cosmic Origins Spectrograph (COS) on board the Hubble Space Telescope, we study the cold ISM properties of the lens galaxy in B1152+199 at a redshift of z=0.4377. Since existing optical extinction and X-ray absorption measurements of the lens have revealed a large amount of cold ISM, we expected to detect a damped Lya absorption (DLA) system in the near ultraviolet spectrum; however, our upper limit on the HI column density is several orders of magnitude below the expectation. We also marginally detect OI and CII absorption lines associated with the lens galaxy in the COS spectrum. Thus, the lens galaxy is identified as a ghostly DLA system, and further investigations of these ghostly DLA systems would be important to characterize the biases of using DLAs to probe the matter density of the universe. Although preliminary, the most likely explanation of the non-detection of the DLA is because of the Lya emission of the lens galaxy that fills in the absorption trough, with a Lya luminosity of 4e42 erg/s.

astro-ph.GA↗