Search

In this paper, we have studied the convectively coupled equatorially trapped waves in rotating stars, with and without magnetic field. The equatorial trapped HD and MHD Poincar\'e, Rossby, mixed Rossby-Poincar\'e, and Kelvin waves were identified. The effects of stratification and non-traditional Coriolis force terms have been investigated. When the flow is strongly stratified, the wave frequencies of the convectively coupled model are almost the same as those of shallow water model. However, when the flow is weakly stratified, the wave frequencies are constrained by the buoyancy frequency. The non-traditional Coriolis terms affect the widths and phases of the equatorial waves. The width increases with the increasing non-traditional Coriolis parameter. Phase shift occurs when the non-traditional Coriolis parameter is included. Magnetic effect is significant when the magnetic field is strong. We have applied the model in the solar atmosphere and solar tachocline to explain the Rieger type periodicities. For the solar atmosphere, when magnetic effect is taken into account, we find that the magnetic field should be smaller than $5G$ in the solar photosphere. Otherwise, the Rieger type periodicities can be only attributed to long Rossby waves. For the solar tachocline, we find that magnetic field of the solar tachocline should be smaller than $50kG$ to observe the 160 days Rieger period. In addition, we find that the effect of the non-traditional Coriolis terms is not obvious in the solar photosphere, but its effect on the tachocline is significant., Comment: published at ApJ

Theory suggests that the orbits of a large fraction of binary systems, including planet-star binary systems, shrink by few orders of magnitude after formation. But so far, only one hot Jupiter with tidally-driven orbital decay has been found by transit timing variations. We propose to search for orbital decay companions in heartbeat star systems because the orbital angular momentum is effectively transferred to the host star causing tidal dissipation. KIC 3766353 is one of the heartbeat stars with tidally excited oscillations. We acquired the primary and the secondary eclipse time variations from the \textit{Kepler} photometric light curves. Timing analysis shows that KIC 3766353 is a hierarchical triple system with a hidden third body and a red dwarf (mass $0.35\ M_{\odot}$, radius $0.34\ R_{\odot}$) in its inner orbit. The minimum mass of the third body is $\sim 0.26 \ M_{\odot}$, and the distance from the inner orbital is $\sim 111.4 \ R_{\odot}$. The period decay rate of the red dwarf is approximately 358 ms yr$^{-1}$. The combined effects of the light-travel time and the orbital decay lead to the observed timing variations. Future monitoring with a long time base-line observations is required to delve into the contributions of these two effects., Comment: 10 pages, 7 figures, and 2 tables. Accepted for publication in MNRAS

We present ReasonBert, a pre-training method that augments language models with the ability to reason over long-range relations and multiple, possibly hybrid contexts. Unlike existing pre-training methods that only harvest learning signals from local contexts of naturally occurring texts, we propose a generalized notion of distant supervision to automatically connect multiple pieces of text and tables to create pre-training examples that require long-range reasoning. Different types of reasoning are simulated, including intersecting multiple pieces of evidence, bridging from one piece of evidence to another, and detecting unanswerable cases. We conduct a comprehensive evaluation on a variety of extractive question answering datasets ranging from single-hop to multi-hop and from text-only to table-only to hybrid that require various reasoning capabilities and show that ReasonBert achieves remarkable improvement over an array of strong baselines. Few-shot experiments further demonstrate that our pre-training method substantially improves sample efficiency., Comment: Accepted to EMNLP'2021. Our code and pre-trained models are available at https://github.com/sunlab-osu/ReasonBERT

Apsidal motion is a gradual shift in the position of periastron. The impact of dynamic tides on apsidal motion has long been debated, because the contribution could not be quantified due to the lack of high quality observations. KIC 4544587 with tidally excited oscillations has been observed by \textit{Kepler} high-precision photometric data based on long time baseline and short-cadence schema. In this paper, we compute the rate of apsidal motion that arises from the dynamic tides as $19.05\pm 1.70$ mrad yr$^{-1}$ via tracking the orbital phase shifts of tidally excited oscillations. We also calculate the procession rate of the orbit due to the Newtonian and general relativistic contribution as $21.49 \pm 2.8$ and $2.4 \pm 0.06$ mrad yr$^{-1}$, respectively. The sum of these three factors is in excellent agreement with the total observational rate of apsidal motion $42.97 \pm 0.18$ mrad yr$^{-1}$ measured by eclipse timing variations. The tidal effect accounts for about 44\% of the overall observed apsidal motion and is comparable to that of the Newtonian term. Dynamic tides have a significant contribution to the apsidal motion. The analysis method mentioned in this paper presents an alternative approach to measuring the contribution of the dynamic tides quantitatively., Comment: 10 pages, 5 figures, accepted by ApJ

State-of-the-art models in natural language processing rely on separate rigid subword tokenization algorithms, which limit their generalization ability and adaptation to new settings. In this paper, we propose a new model inductive bias that learns a subword tokenization end-to-end as part of the model. To this end, we introduce a soft gradient-based subword tokenization module (GBST) that automatically learns latent subword representations from characters in a data-driven fashion. Concretely, GBST enumerates candidate subword blocks and learns to score them in a position-wise fashion using a block scoring network. We additionally introduce Charformer, a deep Transformer model that integrates GBST and operates on the byte level. Via extensive experiments on English GLUE, multilingual, and noisy text datasets, we show that Charformer outperforms a series of competitive byte-level baselines while generally performing on par and sometimes outperforming subword-based models. Additionally, Charformer is fast, improving the speed of both vanilla byte-level and subword-level Transformers by 28%-100% while maintaining competitive quality. We believe this work paves the way for highly performant token-free models that are trained completely end-to-end., Comment: ICLR 2022 Camera Ready

We aim to renew interest in a particular multi-document summarization (MDS) task which we call AgreeSum: agreement-oriented multi-document summarization. Given a cluster of articles, the goal is to provide abstractive summaries that represent information common and faithful to all input articles. Given the lack of existing datasets, we create a dataset for AgreeSum, and provide annotations on article-summary entailment relations for a subset of the clusters in the dataset. We aim to create strong baselines for the task by applying the top-performing pretrained single-document summarization model PEGASUS onto AgreeSum, leveraging both annotated clusters by supervised losses, and unannotated clusters by T5-based entailment-related and language-related losses. Compared to other baselines, both automatic evaluation and human evaluation show better article-summary and cluster-summary entailment in generated summaries. On a separate note, we hope that our article-summary entailment annotations contribute to the community's effort in improving abstractive summarization faithfulness., Comment: Findings of ACL 2021

Effectively modeling text-rich fresh content such as news articles at document-level is a challenging problem. To ensure a content-based model generalize well to a broad range of applications, it is critical to have a training dataset that is large beyond the scale of human labels while achieving desired quality. In this work, we address those two challenges by proposing a novel approach to mine semantically-relevant fresh documents, and their topic labels, with little human supervision. Meanwhile, we design a multitask model called NewsEmbed that alternatively trains a contrastive learning with a multi-label classification to derive a universal document encoder. We show that the proposed approach can provide billions of high quality organic training examples and can be naturally extended to multilingual setting where texts in different languages are encoded in the same semantic space. We experimentally demonstrate NewsEmbed's competitive performance across multiple natural language understanding tasks, both supervised and unsupervised., Comment: Accepted in SIGKDD 2021

Pre-trained text encoders such as BERT and its variants have recently achieved state-of-the-art performances on many NLP tasks. While being effective, these pre-training methods typically demand massive computation resources. To accelerate pre-training, ELECTRA trains a discriminator that predicts whether each input token is replaced by a generator. However, this new task, as a binary classification, is less semantically informative. In this study, we present a new text encoder pre-training method that improves ELECTRA based on multi-task learning. Specifically, we train the discriminator to simultaneously detect replaced tokens and select original tokens from candidate sets. We further develop two techniques to effectively combine all pre-training tasks: (1) using attention-based networks for task-specific heads, and (2) sharing bottom layers of the generator and the discriminator. Extensive experiments on GLUE and SQuAD datasets demonstrate both the effectiveness and the efficiency of our proposed method., Comment: Accepted in Findings of ACL 2021

We present 127 new transit light curves for 39 hot Jupiter systems, obtained over the span of five years by two ground-based telescopes. A homogeneous analysis of these newly collected light curves together with archived spectroscopic, photometric, and Doppler velocimetric data using EXOFASTv2 leads to a significant improvement in the physical and orbital parameters of each system. All of our stellar radii are constrained to accuracies of better than 3\%. The planetary radii for 37 of our 39 targets are determined to accuracies of better than $5\%$. Compared to our results, the literature eccentricities are preferentially overestimated due to the Lucy-Sweeney bias. Our new photometric observations therefore allow for significant improvement in the orbital ephemerides of each system. Our correction of the future transit window amounts to a change exceeding $10\,{\rm min}$ for ten targets at the time of JWST's launch, including a $72\,{\rm min}$ change for WASP-56. The measured transit mid-times for both literature light curves and our new photometry show no significant deviations from the updated linear ephemerides, ruling out in each system the presence of companion planets with masses greater than $0.39 - 5.0\, rm M_{\oplus}$, $1.23 - 14.36\, \rm M_{\oplus}$, $1.65 - 21.18\, \rm M_{\oplus}$, and $0.69 - 6.75\, \rm M_{\oplus}$ near the 1:2, 2:3, 3:2, and 2:1 resonances with the hot Jupiters , respectively, at a confidence level of $\pm 1\,\sigma$. The absence of resonant companion planets in the hot Jupiter systems is inconsistent with the conventional expectation from disk migration., Comment: 28 pages,9 figures, accepted for publication in ApJS

In this paper, we study wave transmission in a rotating fluid with multiple alternating convectively stable and unstable layers. We have discussed wave transmissions in two different circumstances: cases where the wave is propagative in each layer and cases where wave tunneling occurs. We find that efficient wave transmission can be achieved by `resonant propagation' or `resonant tunneling', even when stable layers are strongly stratified, and we call this phenomenon `enhanced wave transmission'. Enhanced wave transmission only occurs when the total number of layers is odd (embedding stable layers are alternatingly embedded within clamping convective layers, or vise versa). For wave propagation, the occurrence of enhanced wave transmission requires that clamping layers have similar properties, the thickness of each clamping layer is close to a multiple of the half wavelength of the corresponding propagative wave, and the total thickness of embedded layers is close to a multiple of the half wavelength of the corresponding propagating wave (resonant propagation). For wave tunneling, we have considered two cases: tunneling of gravity waves and tunneling of inertial waves. In both cases, efficient tunneling requires that clamping layers have similar properties, the thickness of each embedded layer is much smaller than the corresponding e-folding decay distance, and the thickness of each clamping layer is close to a multiple-and-a-half of half wavelength (resonant tunneling)., Comment: accepted to J. Fluid Mech

A large majority of American adults get at least some of their news from the Internet. Even though many online news products have the goal of informing their users about the news, they lack scalable and reliable tools for measuring how well they are achieving this goal, and therefore have to resort to noisy proxy metrics (e.g., click-through rates or reading time) to track their performance. As a first step towards measuring news informedness at a scale, we study the problem of quiz-style multiple-choice question generation, which may be used to survey users about their knowledge of recent news. In particular, we formulate the problem as two sequence-to-sequence tasks: question-answer generation (QAG) and distractor, or incorrect answer, generation (DG). We introduce NewsQuizQA, the first dataset intended for quiz-style question-answer generation, containing 20K human written question-answer pairs from 5K news article summaries. Using this dataset, we propose a series of novel techniques for applying large pre-trained Transformer encoder-decoder models, namely PEGASUS and T5, to the tasks of question-answer generation and distractor generation. We show that our models outperform strong baselines using both automated metrics and human raters. We provide a case study of running weekly quizzes on real-world users via the Google Surveys platform over the course of two months. We found that users generally found the automatically generated questions to be educational and enjoyable. Finally, to serve the research community, we are releasing the NewsQuizQA dataset.

Given a data set, misleading conclusions can be drawn from it by cherry-picking selected samples. One important class of conclusions is a trend derived from a data set of values over time. Our goal is to evaluate whether the 'trends' described by the extracted samples are representative of the true situation represented in the data. We demonstrate MithraDetective, a system to compute a support score to indicate how cherry-picked a statement is; that is, whether the reported trend is well-supported by the data. The system can also be used to discover more supported alternatives. MithraDetective provides an interactive visual interface for both tasks.

In this paper, we study the inertial and gravity wave transmissions near the radiative-convective boundaries in the {\it f}-plane. Two configurations have been considered: waves propagate from the convective layer to the radiative stratified stable layer, or In this paper, we study inertial and gravity wave transmissions near radiative-convective boundaries on the {\it f}-plane. Two configurations have been considered: waves propagate from the convective layer to the radiative stratified stable layer, or the other way around. It has been found that waves prefer to survive at low latitudes when the stable layer is strongly stratified ($N^2/(2\Omega)^2>1$). When the stable layer is weakly stratified ($N^2/(2\Omega)^2<1$), however, waves can survive at any latitude if the meridional wavenumber is large. Then we have discussed transmission ratios for two buoyancy frequency structures: the uniform stratification, and the continuously varying stratification. For the uniform stratification, we have found that the transmission is efficient when the rotation is rapid, or when the wave is near the critical colatitude. For the continuously varying stratification, we have discussed the transmission ratio when the square of buoyancy frequency is an algebraic function $N^2\propto z^{\nu} (\nu >0)$. We have found that the transmission can be efficient when the rotation is rapid, or when the wave is near the critical colatitude, or when the thickness of the stratification layer is far greater than the horizontal wave length. The transmission ratio does not depend on the configurations (radiative layer sits above convective layer, or vice versa; wave propagates outward or inward), but only on characteristics of the wave (frequency and wavenumber) and the fluid (degree of stratification)., Comment: submitted to JFM

Relational tables on the Web store a vast amount of knowledge. Owing to the wealth of such tables, there has been tremendous progress on a variety of tasks in the area of table understanding. However, existing work generally relies on heavily-engineered task-specific features and model architectures. In this paper, we present TURL, a novel framework that introduces the pre-training/fine-tuning paradigm to relational Web tables. During pre-training, our framework learns deep contextualized representations on relational tables in an unsupervised manner. Its universal model design with pre-trained representations can be applied to a wide range of tasks with minimal task-specific fine-tuning. Specifically, we propose a structure-aware Transformer encoder to model the row-column structure of relational tables, and present a new Masked Entity Recovery (MER) objective for pre-training to capture the semantics and knowledge in large-scale unlabeled data. We systematically evaluate TURL with a benchmark consisting of 6 different tasks for table understanding (e.g., relation extraction, cell filling). We show that TURL generalizes well to all tasks and substantially outperforms existing methods in almost all instances., Comment: Accepted to VLDB 2021. Extended version with experiments added during revision. Our source code, benchmark, as well as pre-trained models will be available on https://github.com/sunlab-osu/TURL

We evaluate the performance of transformer encoders with various decoders for information organization through a new task: generation of section headings for Wikipedia articles. Our analysis shows that decoders containing attention mechanisms over the encoder output achieve high-scoring results by generating extractive text. In contrast, a decoder without attention better facilitates semantic encoding and can be used to generate section embeddings. We additionally introduce a new loss function, which further encourages the decoder to generate high-quality embeddings., Comment: Accepted to WNGT Workshop at ACL 2020

The advent of natural language understanding (NLU) benchmarks for English, such as GLUE and SuperGLUE allows new NLU models to be evaluated across a diverse set of tasks. These comprehensive benchmarks have facilitated a broad range of research and applications in natural language processing (NLP). The problem, however, is that most such benchmarks are limited to English, which has made it difficult to replicate many of the successes in English NLU for other languages. To help remedy this issue, we introduce the first large-scale Chinese Language Understanding Evaluation (CLUE) benchmark. CLUE is an open-ended, community-driven project that brings together 9 tasks spanning several well-established single-sentence/sentence-pair classification tasks, as well as machine reading comprehension, all on original Chinese text. To establish results on these tasks, we report scores using an exhaustive set of current state-of-the-art pre-trained Chinese models (9 in total). We also introduce a number of supplementary datasets and additional tools to help facilitate further progress on Chinese NLU. Our benchmark is released at https://www.CLUEbenchmarks.com, Comment: Accepted by COLING2020; 10 pages, 4 figures

In this paper, we investigate the magnetohydrodynamical structure of a jet powered by a spinning black hole, where electromagnetic fields and fluid motion are governed by the Grad-Shafranov equation and the Bernoulli equation, respectively. Assuming steady and axisymmetric jet structure, the global solution is uniquely determined with prescribed plasma loading into the jet and the poloidal shape of the outmost magnetic field line. We apply this model to the jet in the center of nearby radio galaxy M87, and we find it can naturally explain the slow flow acceleration and the flow velocity stratification within $10^5$ gravitational radii from the central black hole. In particular, we find the extremal black hole spin is disfavored by the flow velocity measurements, if the plasma loading to the jet is dominated by the electron/positron pair production at the jet base., Comment: 15 pages, 9 figures, accepted by ApJ

We study the angular momentum transport inside the hot Jupiters under the the influences of gravitational and thermal forcing. Due to the strong stellar irradiation, radiative region develops on top of the convective region. Internal gravity waves are launched at the radiative-convective boundaries (RCBs). The thermal response is dynamical and plays an important role in the angular momentum transport. By separating the gravitational and thermal forcing terms, we identify the thermal effects for increasing the angular momentum transport. For the low frequency (in the co-rotating frame with planets) prograde (retrograde) tidal frequency, the angular momentum flux is positive (negative). The tidal interactions tends to drive the planet to the synchronous state. We find that the angular momentum transport associated with the internal gravity wave is very sensitive to relative position between the RCB and the penetration depth of the thermal forcing. If the RCB is in the vicinity of the thermal forcing penetration depth, even with small amplitude thermal forcing, the thermally driven angular momentum flux could be much larger than the flux induced by gravitational forcing. The thermally enhanced torque could drive the planet to the synchronous state in as short as a few $10^4$ years., Comment: 6 pages, 3 figures, ApJ Letter Accepted

Millions of news articles are published online every day, which can be overwhelming for readers to follow. Grouping articles that are reporting the same event into news stories is a common way of assisting readers in their news consumption. However, it remains a challenging research problem to efficiently and effectively generate a representative headline for each story. Automatic summarization of a document set has been studied for decades, while few studies have focused on generating representative headlines for a set of articles. Unlike summaries, which aim to capture most information with least redundancy, headlines aim to capture information jointly shared by the story articles in short length, and exclude information that is too specific to each individual article. In this work, we study the problem of generating representative headlines for news stories. We develop a distant supervision approach to train large-scale generation models without any human annotation. This approach centers on two technical components. First, we propose a multi-level pre-training framework that incorporates massive unlabeled corpus with different quality-vs.-quantity balance at different levels. We show that models trained within this framework outperform those trained with pure human curated corpus. Second, we propose a novel self-voting-based article attention layer to extract salient information shared by multiple articles. We show that models that incorporate this layer are robust to potential noises in news stories and outperform existing baselines with or without noises. We can further enhance our model by incorporating human labels, and we show our distant supervision approach significantly reduces the demand on labeled data., Comment: WebConf 2020 (WWW 2020)

In this paper, we introduce the NER dataset from CLUE organization (CLUENER2020), a well-defined fine-grained dataset for named entity recognition in Chinese. CLUENER2020 contains 10 categories. Apart from common labels like person, organization, and location, it contains more diverse categories. It is more challenging than current other Chinese NER datasets and could better reflect real-world applications. For comparison, we implement several state-of-the-art baselines as sequence labeling tasks and report human performance, as well as its analysis. To facilitate future work on fine-grained NER for Chinese, we release our dataset, baselines, and leader-board., Comment: 6 pages, 5 tables, 1 figure