arXiv subjects

Xinyi He

Publications and source records attributed to Xinyi He.

26 records · Page 2Linked to original sources

arXiv2022

Towards Robust Numerical Question Answering: Diagnosing Numerical Capabilities of NLP Systems

Numerical Question Answering is the task of answering questions that require numerical capabilities. Previous works introduce general adversarial attacks to Numerical Question Answering, while not systematically exploring numerical capabilities specific to the topic. In this paper, we propose to conduct numerical capability diagnosis on a series of Numerical Question Answering systems and datasets. A series of numerical capabilities are highlighted, and corresponding dataset perturbations are designed. Empirical results indicate that existing systems are severely challenged by these perturbations. E.g., Graph2Tree experienced a 53.83% absolute accuracy drop against the ``Extra'' perturbation on ASDiv-a, and BART experienced 13.80% accuracy drop against the ``Language'' perturbation on the numerical subset of DROP. As a counteracting approach, we also investigate the effectiveness of applying perturbations as data augmentation to relieve systems' lack of robust numerical capabilities. With experiment analysis and empirical studies, it is demonstrated that Numerical Question Answering with robust numerical capabilities is still to a large extent an open question. We discuss future directions of Numerical Question Answering and summarize guidelines on future dataset collection and system design.

cs.CL↗

arXiv2022

ASTA: Learning Analytical Semantics over Tables for Intelligent Data Analysis and Visualization

Intelligent analysis and visualization of tables use techniques to automatically recommend useful knowledge from data, thus freeing users from tedious multi-dimension data mining. While many studies have succeeded in automating recommendations through rules or machine learning, it is difficult to generalize expert knowledge and provide explainable recommendations. In this paper, we present the recommendation of conditional formatting for the first time, together with chart recommendation, to exemplify intelligent table analysis. We propose analytical semantics over tables to uncover common analysis pattern behind user-created analyses. Here, we design analytical semantics by separating data focus from user intent, which extract the user motivation from data and human perspective respectively. Furthermore, the ASTA framework is designed by us to apply analytical semantics to multiple automated recommendations. ASTA framework extracts data features by designing signatures based on expert knowledge, and enables data referencing at field- (chart) or cell-level (conditional formatting) with pre-trained models. Experiments show that our framework achieves recall at top 1 of 62.86% on public chart corpora, outperforming the best baseline about 14%, and achieves 72.31% on the collected corpus ConFormT, validating that ASTA framework is effective in providing accurate and explainable recommendations.

cs.DB↗

arXiv2022

Table Pre-training: A Survey on Model Architectures, Pre-training Objectives, and Downstream Tasks

Since a vast number of tables can be easily collected from web pages, spreadsheets, PDFs, and various other document types, a flurry of table pre-training frameworks have been proposed following the success of text and images, and they have achieved new state-of-the-arts on various tasks such as table question answering, table type recognition, column relation classification, table search, formula prediction, etc. To fully use the supervision signals in unlabeled tables, a variety of pre-training objectives have been designed and evaluated, for example, denoising cell values, predicting numerical relationships, and implicitly executing SQLs. And to best leverage the characteristics of (semi-)structured tables, various tabular language models, particularly with specially-designed attention mechanisms, have been explored. Since tables usually appear and interact with free-form text, table pre-training usually takes the form of table-text joint pre-training, which attracts significant research interests from multiple domains. This survey aims to provide a comprehensive review of different model designs, pre-training objectives, and downstream tasks for table pre-training, and we further share our thoughts and vision on existing challenges and future opportunities.

cs.CL↗

arXiv2022

Pressure-induced superconductivity in kagome single crystal Pd3P2S8

Kagome lattice offers unique opportunities for the exploration of unusual quantum states of correlated electrons. Here, we report on the observation of superconductivity in a kagome single crystal Pd3P2S8 when a semiconducting to metallic transition is driven by pressure. High-pressure resistance measurements show that the metallization and superconductivity are simultaneously observed at about 11 GPa. With increasing pressure, the superconducting critical temperature Tc is monotonously enhanced from 2.6 K to a maximum 7.7 K at ~52 GPa. Interestingly, superconductivity retains when the pressure is fully released. Synchrotron XRD and Raman experiments consistently evidence that the emergence of superconductivity is accompanied with an amorphization and the retainability of superconductivity upon decompression can be attributed to the irreversibility of the amorphization.

cond-mat.supr-con↗

arXiv2022

On an error term for the first moment of twisted $L$-functions

Let $f$ be a Hecke-Maass cusp form for the full modular group and let $χ$ be a primitive Dirichlet character modulo a prime $q$. Let $s_0=σ_0+it_0$ with $\frac{1}{2}\leqσ_0<1$. We improve the error term for the first moment of $L(s_0,f\otimesχ)\overline{L(s_0,χ)}$ over the family of even primitive Dirichlet characters. As an application, we show that for any $t\in\mathbb{R}$, there exists a primitive Dirichlet character $χ$ modulo $q$ for which $L(1/2+it,f\otimesχ)L(1/2+it,χ)\neq0$ if the prime $q$ satisfies $q\gg (1+|t|)^{\frac{543}{25}+\varepsilon}$.

math.NT↗

arXiv2021

MultiVision: Designing Analytical Dashboards with Deep Learning Based Recommendation

We contribute a deep-learning-based method that assists in designing analytical dashboards for analyzing a data table. Given a data table, data workers usually need to experience a tedious and time-consuming process to select meaningful combinations of data columns for creating charts. This process is further complicated by the need of creating dashboards composed of multiple views that unveil different perspectives of data. Existing automated approaches for recommending multiple-view visualizations mainly build on manually crafted design rules, producing sub-optimal or irrelevant suggestions. To address this gap, we present a deep learning approach for selecting data columns and recommending multiple charts. More importantly, we integrate the deep learning models into a mixed-initiative system. Our model could make recommendations given optional user-input selections of data columns. The model, in turn, learns from provenance data of authoring logs in an offline manner. We compare our deep learning model with existing methods for visualization recommendation and conduct a user study to evaluate the usefulness of the system.

cs.HC↗

arXiv2021

Table2Charts: Recommending Charts by Learning Shared Table Representations

It is common for people to create different types of charts to explore a multi-dimensional dataset (table). However, to recommend commonly composed charts in real world, one should take the challenges of efficiency, imbalanced data and table context into consideration. In this paper, we propose Table2Charts framework which learns common patterns from a large corpus of (table, charts) pairs. Based on deep Q-learning with copying mechanism and heuristic searching, Table2Charts does table-to-sequence generation, where each sequence follows a chart template. On a large spreadsheet corpus with 165k tables and 266k charts, we show that Table2Charts could learn a shared representation of table fields so that recommendation tasks on different chart types could mutually enhance each other. Table2Charts outperforms other chart recommendation systems in both multi-type task (with doubled recall numbers R@3=0.61 and R@1=0.43) and human evaluations.

cs.DB↗

arXiv2020

Crystal structure built from a GeO$_6$-GeO$_5$ polyhedra network with high thermal stability: $β$-SrGe$_2$O$_5$

By tackling the challenge of extending transparent oxide semiconductors to Ge based oxides, we have found a not-yet-reported crystal structure, named $β$-SrGe$_2$O$_5$, which is composed of edge-sharing GeO$_6$ octahedra interconnected by GeO$_5$ bipyramids. Single crystals were successfully grown by the high-pressure flux method. $β$-SrGe$_2$O$_5$ has a band gap of 5.2 eV and a dispersive conduction band with an effective mass as small as 0.34 times the electron rest mass, which originates from the edge-sharing GeO$_6$ octahedra network. Although known compounds with octahedral GeO$_6$ coordination are commonly unstable at atmospheric pressure and elevated temperatures, $β$-SrGe$_2$O$_5$ exhibits thermal stability up to 700 $^\circ$C.

cond-mat.mtrl-sci↗