arXiv subjects

Yu Su

Publications and source records attributed to Yu Su.

180 records · Page 10Linked to original sources

arXiv2020

Logical Natural Language Generation from Open-Domain Tables

Neural natural language generation (NLG) models have recently shown remarkable progress in fluency and coherence. However, existing studies on neural NLG are primarily focused on surface-level realizations with limited emphasis on logical inference, an important aspect of human thinking and language. In this paper, we suggest a new NLG task where a model is tasked with generating natural language statements that can be \emph{logically entailed} by the facts in an open-domain semi-structured table. To facilitate the study of the proposed logical NLG problem, we use the existing TabFact dataset \cite{chen2019tabfact} featured with a wide range of logical/symbolic inferences as our testbed, and propose new automatic metrics to evaluate the fidelity of generation models w.r.t.\ logical inference. The new task poses challenges to the existing monotonic generation frameworks due to the mismatch between sequence order and logical order. In our experiments, we comprehensively survey different generation architectures (LSTM, Transformer, Pre-Trained LM) trained with different algorithms (RL, Adversarial Training, Coarse-to-Fine) on the dataset and made following observations: 1) Pre-Trained LM can significantly boost both the fluency and logical fidelity metrics, 2) RL and Adversarial Training are trading fluency for fidelity, 3) Coarse-to-Fine generation can help partially alleviate the fidelity issue while maintaining high language fluency. The code and data are available at \url{https://github.com/wenhuchen/LogicNLG}.

cs.CL↗

arXiv2019

Decision Propagation Networks for Image Classification

High-level (e.g., semantic) features encoded in the latter layers of convolutional neural networks are extensively exploited for image classification, leaving low-level (e.g., color) features in the early layers underexplored. In this paper, we propose a novel Decision Propagation Module (DPM) to make an intermediate decision that could act as category-coherent guidance extracted from early layers, and then propagate it to the latter layers. Therefore, by stacking a collection of DPMs into a classification network, the generated Decision Propagation Network is explicitly formulated as to progressively encode more discriminative features guided by the decision, and then refine the decision based on the new generated features layer by layer. Comprehensive results on four publicly available datasets validate DPM could bring significant improvements for existing classification networks with minimal additional computational cost and is superior to the state-of-the-art methods.

cs.CV↗

arXiv2019

Model-based Interactive Semantic Parsing: A Unified Framework and A Text-to-SQL Case Study

As a promising paradigm, interactive semantic parsing has shown to improve both semantic parsing accuracy and user confidence in the results. In this paper, we propose a new, unified formulation of the interactive semantic parsing problem, where the goal is to design a model-based intelligent agent. The agent maintains its own state as the current predicted semantic parse, decides whether and where human intervention is needed, and generates a clarification question in natural language. A key part of the agent is a world model: it takes a percept (either an initial question or subsequent feedback from the user) and transitions to a new state. We then propose a simple yet remarkably effective instantiation of our framework, demonstrated on two text-to-SQL datasets (WikiSQL and Spider) with different state-of-the-art base semantic parsers. Compared to an existing interactive semantic parsing approach that treats the base parser as a black box, our approach solicits less user feedback but yields higher run-time accuracy.

cs.CL↗

arXiv2019

EKT: Exercise-aware Knowledge Tracing for Student Performance Prediction

For offering proactive services to students in intelligent education, one of the fundamental tasks is predicting their performance (e.g., scores) on future exercises, where it is necessary to track each student's knowledge acquisition during her exercising activities. However, existing approaches can only exploit the exercising records of students, and the problem of extracting rich information existed in the exercise's materials (e.g., knowledge concepts, exercise content) to achieve both precise predictions of student performance and interpretable analysis of knowledge acquisition remains underexplored. In this paper, we present a holistic study of student performance prediction. To directly achieve the primary goal of prediction, we first propose a general Exercise-Enhanced Recurrent Neural Network (EERNN) framework by exploring both student's records and the exercise contents. In EERNN, we simply summarize each student's state into an integrated vector and trace it with a recurrent neural network, where we design a bidirectional LSTM to learn the encoding of each exercise's content. For making predictions, we propose two implementations under EERNN with different strategies, i.e., EERNNM with Markov property and EERNNA with Attention mechanism. Then, to explicitly track student's knowledge acquisition on multiple knowledge concepts, we extend EERNN to an explainable Exercise-aware Knowledge Tracing (EKT) by incorporating the knowledge concept effects, where the student's integrated state vector is extended to a knowledge state matrix. In EKT, we further develop a memory network for quantifying how much each exercise can affect the mastery of students on concepts during the exercising process. Finally, we conduct extensive experiments on large-scale real-world data. The results demonstrate the prediction effectiveness of two frameworks as well as the superior interpretability of EKT.

cs.CY↗

arXiv2019

Global Textual Relation Embedding for Relational Understanding

Pre-trained embeddings such as word embeddings and sentence embeddings are fundamental tools facilitating a wide range of downstream NLP tasks. In this work, we investigate how to learn a general-purpose embedding of textual relations, defined as the shortest dependency path between entities. Textual relation embedding provides a level of knowledge between word/phrase level and sentence level, and we show that it can facilitate downstream tasks requiring relational understanding of the text. To learn such an embedding, we create the largest distant supervision dataset by linking the entire English ClueWeb09 corpus to Freebase. We use global co-occurrence statistics between textual and knowledge base relations as the supervision signal to train the embedding. Evaluation on two relational understanding tasks demonstrates the usefulness of the learned textual relation embedding. The data and code can be found at https://github.com/czyssrs/GloREPlus

cs.CL↗

arXiv2019

QuesNet: A Unified Representation for Heterogeneous Test Questions

Understanding learning materials (e.g. test questions) is a crucial issue in online learning systems, which can promote many applications in education domain. Unfortunately, many supervised approaches suffer from the problem of scarce human labeled data, whereas abundant unlabeled resources are highly underutilized. To alleviate this problem, an effective solution is to use pre-trained representations for question understanding. However, existing pre-training methods in NLP area are infeasible to learn test question representations due to several domain-specific characteristics in education. First, questions usually comprise of heterogeneous data including content text, images and side information. Second, there exists both basic linguistic information as well as domain logic and knowledge. To this end, in this paper, we propose a novel pre-training method, namely QuesNet, for comprehensively learning question representations. Specifically, we first design a unified framework to aggregate question information with its heterogeneous inputs into a comprehensive vector. Then we propose a two-level hierarchical pre-training algorithm to learn better understanding of test questions in an unsupervised way. Here, a novel holed language model objective is developed to extract low-level linguistic features, and a domain-oriented objective is proposed to learn high-level logic and knowledge. Moreover, we show that QuesNet has good capability of being fine-tuned in many question-based tasks. We conduct extensive experiments on large-scale real-world question data, where the experimental results clearly demonstrate the effectiveness of QuesNet for question understanding as well as its superior applicability.

cs.LG↗

arXiv2019

How Large a Vocabulary Does Text Classification Need? A Variational Approach to Vocabulary Selection

With the rapid development in deep learning, deep neural networks have been widely adopted in many real-life natural language applications. Under deep neural networks, a pre-defined vocabulary is required to vectorize text inputs. The canonical approach to select pre-defined vocabulary is based on the word frequency, where a threshold is selected to cut off the long tail distribution. However, we observed that such simple approach could easily lead to under-sized vocabulary or over-sized vocabulary issues. Therefore, we are interested in understanding how the end-task classification accuracy is related to the vocabulary size and what is the minimum required vocabulary size to achieve a specific performance. In this paper, we provide a more sophisticated variational vocabulary dropout (VVD) based on variational dropout to perform vocabulary selection, which can intelligently select the subset of the vocabulary to achieve the required performance. To evaluate different algorithms on the newly proposed vocabulary selection problem, we propose two new metrics: Area Under Accuracy-Vocab Curve and Vocab Size under X\% Accuracy Drop. Through extensive experiments on various NLP classification tasks, our variational framework is shown to significantly outperform the frequency-based and other selection baselines on these metrics.

cs.CL↗

arXiv2018

Learning to Compose Topic-Aware Mixture of Experts for Zero-Shot Video Captioning

Although promising results have been achieved in video captioning, existing models are limited to the fixed inventory of activities in the training corpus, and do not generalize to open vocabulary scenarios. Here we introduce a novel task, zero-shot video captioning, that aims at describing out-of-domain videos of unseen activities. Videos of different activities usually require different captioning strategies in many aspects, i.e. word selection, semantic construction, and style expression etc, which poses a great challenge to depict novel activities without paired training data. But meanwhile, similar activities share some of those aspects in common. Therefore, We propose a principled Topic-Aware Mixture of Experts (TAMoE) model for zero-shot video captioning, which learns to compose different experts based on different topic embeddings, implicitly transferring the knowledge learned from seen activities to unseen ones. Besides, we leverage external topic-related text corpus to construct the topic embedding for each activity, which embodies the most relevant semantic vectors within the topic. Empirical results not only validate the effectiveness of our method in utilizing semantic knowledge for video captioning, but also show its strong generalization ability when describing novel activities.

cs.CL↗

arXiv2018

Algorithms for Optimal Control with Fixed-Rate Feedback

We consider a discrete-time linear quadratic Gaussian networked control setting where the (full information) observer and controller are separated by a fixed-rate noiseless channel. The minimal rate required to stabilize such a system has been well studied. However, for a given fixed rate, how to quantize the states so as to optimize performance is an open question of great theoretical and practical significance. We concentrate on minimizing the control cost for first-order scalar systems. To that end, we use the Lloyd-Max algorithm and leverage properties of logarithmically-concave functions and sequential Bayesian filtering to construct the optimal quantizer that greedily minimizes the cost at every time instant. By connecting the globally optimal scheme to the problem of scalar successive refinement, we argue that its gain over the proposed greedy algorithm is negligible. This is significant since the globally optimal scheme is often computationally intractable. All the results are proven for the more general case of disturbances with logarithmically-concave distributions and rate-limited time-varying noiseless channels. We further extend the framework to event-triggered control by allowing to convey information via an additional "silent symbol", i.e., by avoiding transmitting bits; by constraining the minimal probability of silence we attain a tradeoff between the transmission rate and the control cost for rates below one bit per sample.

eess.SY↗

arXiv2018

XL-NBT: A Cross-lingual Neural Belief Tracking Framework

Task-oriented dialog systems are becoming pervasive, and many companies heavily rely on them to complement human agents for customer service in call centers. With globalization, the need for providing cross-lingual customer support becomes more urgent than ever. However, cross-lingual support poses great challenges---it requires a large amount of additional annotated data from native speakers. In order to bypass the expensive human annotation and achieve the first step towards the ultimate goal of building a universal dialog system, we set out to build a cross-lingual state tracking framework. Specifically, we assume that there exists a source language with dialog belief tracking annotations while the target languages have no annotated dialog data of any form. Then, we pre-train a state tracker for the source language as a teacher, which is able to exploit easy-to-access parallel data. We then distill and transfer its own knowledge to the student state tracker in target languages. We specifically discuss two types of common parallel resources: bilingual corpus and bilingual dictionary, and design different transfer learning strategies accordingly. Experimentally, we successfully use English state tracker as the teacher to transfer its knowledge to both Italian and German trackers and achieve promising results.

cs.CL↗

arXiv2018

Elliptic problem involving finite many critical exponents in $\mathbb{R}^{N}$

In this paper, we consider the following problem $$ -Δu -ζ\frac{u}{|x|^{2}} = \sum_{i=1}^{k} \left( \int_{\mathbb{R}^{N}} \frac{|u|^{2^{*}_{α_{i}}}}{|x-y|^{α_{i}}} \mathrm{d}y \right) |u|^{2^{*}_{α_{i}}-2}u + |u|^{2^{*}-2}u , \mathrm{~in~} \mathbb{R}^{N}, $$ where $N\geqslant3$, $ζ\in(0,\frac{(N-2)^{2}}{4})$, $2^{*}=\frac{2N}{N-2}$ is the critical Sobolev exponent, and $2^{*}_{α_{i}}=\frac{2N-α_{i}}{N-2}$ ($i=1,\ldots,k$) are the critical Hardy--Littlewood--Sobolev upper exponents. The parameters $α_{i}$ ($i=1,\ldots,k$) satisfy some suitable assumptions. By using Coulomb--Sobolev space, endpoint refined Sobolev inequality and variational methods, we establish the existence of nontrivial solutions. Our result generalizes the result obtained by Yang and Wu [Adv. Nonlinear Stud. (2017)].

math.AP↗

arXiv2018

Finite many critical problems involving Fractional Laplacians in $\mathbb{R}^{N}$

In this paper, we consider the nonlocal elliptic problems in $\mathbb{R}^{N}$, which involve finite many critical exponents. By using endpoint refined Hardy--Sobolev inequality, fractional Coulomb--Sobolev space and variational method, we establish the existence of nonnegative solution. Our results generalize some results obtained by Yang and Wu [Adv. Nonlinear Stud. (2017) \cite{Yang2017}].

math.AP↗

arXiv2018

The minimizing problem involving p--Laplacian and Hardy--Littlewood--Sobolev upper critical exponent

In this paper, we study the minimizing problem: $$ S_{p,1,α,μ}:= \inf_{u\in W^{1,p}(\mathbb{R}^{N})\setminus\{0\}} \frac{ \int_{\mathbb{R}^{N}}|\nabla u|^{p}\mathrm{d}x - μ\int_{\mathbb{R}^{N}} \frac{|u|^{p}}{|x|^{p}} \mathrm{d}x} {\left( \int_{\mathbb{R}^{N}} \int_{\mathbb{R}^{N}} \frac{|u(x)|^{p^{*}_α}|u(y)|^{p^{*}_α}}{|x-y|^α} \mathrm{d}x \mathrm{d}y \right)^{\frac{p}{2\cdot p^{*}_α}}}, $$ where $N\geqslant3$, $p\in(1,N)$, $μ\in \left[ 0, \left( \frac{N-p}{p} \right)^{p} \right)$, $α\in(0,N)$ and $p^{*}_α= \frac{p}{2}\left(\frac{2N-α}{N-p}\right)$ is the Hardy--Littlewood--Sobolev upper critical exponent. Firstly, by using refinement of Hardy-Littlewood-Sobolev inequality, we prove that $S_{p,1,α,μ}$ is achieved in $\mathbb{R}^{N}$ by a radially symmetric, nonincreasing and nonnegative function. Secondly, we give a estimation of extremal function.

math.AP↗

arXiv2018

Fractional Elliptic problem with Finite many critical Hardy--Sobolev Exponents

In this paper, we consider the following problem: $$ (-Δ)^{s} u -\frac{ζu}{|x|^{2s}} = \sum_{i=1}^{k} \frac{|u|^{2^{*}_{s,θ_{i}}-2}u} {|x|^{θ_{i}}} , \mathrm{~in~} \mathbb{R}^{N}, $$ where $N\geqslant3$, $s\in(0,1)$, $ζ\in \left[ 0,4^{s}\frac{Γ(\frac{N+2s}{4})}{Γ(\frac{N-2s}{4})} \right)$, $2^{*}_{s,θ_{i}}=\frac{2(N-θ_{i})}{N-2s}$ are the critical Hardy--Sobolev exponents, the parameters $θ_{i}$ satisfy a suitable assumption. By using Morrey space, refinement of Hardy--Sobolev inequality and variational method, we establish the existence of nonnegative solution. Our result generalizes the result obtained by Chen [Electronic J. Differ. Eq. (2018) 1--12].

math.AP↗

arXiv2018

Global Relation Embedding for Relation Extraction

We study the problem of textual relation embedding with distant supervision. To combat the wrong labeling problem of distant supervision, we propose to embed textual relations with global statistics of relations, i.e., the co-occurrence statistics of textual and knowledge base relations collected from the entire corpus. This approach turns out to be more robust to the training noise introduced by distant supervision. On a popular relation extraction dataset, we show that the learned textual relation embedding can be used to augment existing relation extraction models and significantly improve their performance. Most remarkably, for the top 1,000 relational facts discovered by the best existing model, the precision can be improved from 83.9% to 89.3%.

cs.CL↗

arXiv2018

Aggregated Channels Network for Real-Time Pedestrian Detection

Convolutional neural networks (CNNs) have demonstrated their superiority in numerous computer vision tasks, yet their computational cost results prohibitive for many real-time applications such as pedestrian detection which is usually performed on low-consumption hardware. In order to alleviate this drawback, most strategies focus on using a two-stage cascade approach. Essentially, in the first stage a fast method generates a significant but reduced amount of high quality proposals that later, in the second stage, are evaluated by the CNN. In this work, we propose a novel detection pipeline that further benefits from the two-stage cascade strategy. More concretely, the enriched and subsequently compressed features used in the first stage are reused as the CNN input. As a consequence, a simpler network architecture, adapted for such small input sizes, allows to achieve real-time performance and obtain results close to the state-of-the-art while running significantly faster without the use of GPU. In particular, considering that the proposed pipeline runs in frame rate, the achieved performance is highly competitive. We furthermore demonstrate that the proposed pipeline on itself can serve as an effective proposal generator.

cs.CV↗

arXiv2017

Cross-domain Semantic Parsing via Paraphrasing

Existing studies on semantic parsing mainly focus on the in-domain setting. We formulate cross-domain semantic parsing as a domain adaptation problem: train a semantic parser on some source domains and then adapt it to the target domain. Due to the diversity of logical forms in different domains, this problem presents unique and intriguing challenges. By converting logical forms into canonical utterances in natural language, we reduce semantic parsing to paraphrasing, and develop an attentive sequence-to-sequence paraphrase model that is general and flexible to adapt to different domains. We discover two problems, small micro variance and large macro variance, of pre-trained word embeddings that hinder their direct use in neural networks, and propose standardization techniques as a remedy. On the popular Overnight dataset, which contains eight domains, we show that both cross-domain training and standardized pre-trained word embeddings can bring significant improvement.

cs.CL↗

arXiv2010

Visual-hint Boundary to Segment Algorithm for Image Segmentation

Image segmentation has been a very active research topic in image analysis area. Currently, most of the image segmentation algorithms are designed based on the idea that images are partitioned into a set of regions preserving homogeneous intra-regions and inhomogeneous inter-regions. However, human visual intuition does not always follow this pattern. A new image segmentation method named Visual-Hint Boundary to Segment (VHBS) is introduced, which is more consistent with human perceptions. VHBS abides by two visual hint rules based on human perceptions: (i) the global scale boundaries tend to be the real boundaries of the objects; (ii) two adjacent regions with quite different colors or textures tend to result in the real boundaries between them. It has been demonstrated by experiments that, compared with traditional image segmentation method, VHBS has better performance and also preserves higher computational efficiency.

cs.CV↗