Models, code, and papers for "Jianfeng Gao":

Challenges in Building Intelligent Open-domain Dialog Systems

May 13, 2019
Minlie Huang, Xiaoyan Zhu, Jianfeng Gao

There is a resurgent interest in developing intelligent open-domain dialog systems due to the availability of large amounts of conversational data and the recent progress on neural approaches to conversational AI. Unlike traditional task-oriented bots, an open-domain dialog system aims to establish long-term connections with users by satisfying the human need for communication, affection, and social belonging. This paper reviews the recent works on neural approaches that are devoted to addressing three challenges in developing such systems: semantics, consistency, and interactiveness. Semantics requires a dialog system to not only understand the content of the dialog but also identify user's social needs during the conversation. Consistency requires the system to demonstrate a consistent personality to win users trust and gain their long-term confidence. Interactiveness refers to the system's ability to generate interpersonal responses to achieve particular social goals such as entertainment, conforming, and task completion. The works we select to present here is based on our unique views and are by no means complete. Nevertheless, we hope that the discussion will inspire new research in developing more intelligent dialog systems.

  Click for Model/Code and Paper
Neural Approaches to Conversational AI

Sep 21, 2018
Jianfeng Gao, Michel Galley, Lihong Li

The present paper surveys neural approaches to conversational AI that have been developed in the last few years. We group conversational systems into three categories: (1) question answering agents, (2) task-oriented dialogue agents, and (3) chatbots. For each category, we present a review of state-of-the-art neural approaches, draw the connection between them and traditional approaches, and discuss the progress that has been made and challenges still being faced, using specific systems and models as case studies.

* Submitted to Foundations and Trends in Information Retrieval (85 pages) 

  Click for Model/Code and Paper
Stochastic Answer Networks for Natural Language Inference

Apr 21, 2018
Xiaodong Liu, Kevin Duh, Jianfeng Gao

We propose a stochastic answer network (SAN) to explore multi-step inference strategies in Natural Language Inference. Rather than directly predicting the results given the inputs, the model maintains a state and iteratively refines its predictions. Our experiments show that SAN achieves the state-of-the-art results on three benchmarks: Stanford Natural Language Inference (SNLI) dataset, MultiGenre Natural Language Inference (MultiNLI) dataset and Quora Question Pairs dataset.

* 5 pages, 1 figures 

  Click for Model/Code and Paper
A Hybrid Neural Network Model for Commonsense Reasoning

Jul 27, 2019
Pengcheng He, Xiaodong Liu, Weizhu Chen, Jianfeng Gao

This paper proposes a hybrid neural network (HNN) model for commonsense reasoning. An HNN consists of two component models, a masked language model and a semantic similarity model, which share a BERT-based contextual encoder but use different model-specific input and output layers. HNN obtains new state-of-the-art results on three classic commonsense reasoning tasks, pushing the WNLI benchmark to 89%, the Winograd Schema Challenge (WSC) benchmark to 75.1%, and the PDP60 benchmark to 90.0%. An ablation study shows that language models and semantic similarity models are complementary approaches to commonsense reasoning, and HNN effectively combines the strengths of both. The code and pre-trained models will be publicly available at

* 9 pages, 3 figures, 6 tables 

  Click for Model/Code and Paper
Improving Multi-Task Deep Neural Networks via Knowledge Distillation for Natural Language Understanding

Apr 20, 2019
Xiaodong Liu, Pengcheng He, Weizhu Chen, Jianfeng Gao

This paper explores the use of knowledge distillation to improve a Multi-Task Deep Neural Network (MT-DNN) (Liu et al., 2019) for learning text representations across multiple natural language understanding tasks. Although ensemble learning can improve model performance, serving an ensemble of large DNNs such as MT-DNN can be prohibitively expensive. Here we apply the knowledge distillation method (Hinton et al., 2015) in the multi-task learning setting. For each task, we train an ensemble of different MT-DNNs (teacher) that outperforms any single model, and then train a single MT-DNN (student) via multi-task learning to \emph{distill} knowledge from these ensemble teachers. We show that the distilled MT-DNN significantly outperforms the original MT-DNN on 7 out of 9 GLUE tasks, pushing the GLUE benchmark (single model) to 83.7\% (1.5\% absolute improvement\footnote{ Based on the GLUE leaderboard at as of April 1, 2019.}). The code and pre-trained models will be made publicly available at

* 8 pages, 2 figures and 3 tables 

  Click for Model/Code and Paper
Multi-Task Deep Neural Networks for Natural Language Understanding

Jan 31, 2019
Xiaodong Liu, Pengcheng He, Weizhu Chen, Jianfeng Gao

In this paper, we present a Multi-Task Deep Neural Network (MT-DNN) for learning representations across multiple natural language understanding (NLU) tasks. MT-DNN not only leverages large amounts of cross-task data, but also benefits from a regularization effect that leads to more general representations in order to adapt to new tasks and domains. MT-DNN extends the model proposed in Liu et al. (2015) by incorporating a pre-trained bidirectional transformer language model, known as BERT (Devlin et al., 2018). MT-DNN obtains new state-of-the-art results on ten NLU tasks, including SNLI, SciTail, and eight out of nine GLUE tasks, pushing the GLUE benchmark to 82.2% (1.8% absolute improvement). We also demonstrate using the SNLI and SciTail datasets that the representations learned by MT-DNN allow domain adaptation with substantially fewer in-domain labels than the pre-trained BERT representations. Our code and pre-trained models will be made publicly available.

* 10 pages, 2 figures and 5 tables 

  Click for Model/Code and Paper
Joint Learning of Distributed Representations for Images and Texts

Apr 28, 2015
Xiaodong He, Rupesh Srivastava, Jianfeng Gao, Li Deng

This technical report provides extra details of the deep multimodal similarity model (DMSM) which was proposed in (Fang et al. 2015, arXiv:1411.4952). The model is trained via maximizing global semantic similarity between images and their captions in natural language using the public Microsoft COCO database, which consists of a large set of images and their corresponding captions. The learned representations attempt to capture the combination of various visual concepts and cues.

* This is a previous tech report of a part of the work of arXiv:1411.4952. In order to avoid confusion, we'd like to withdraw this report from arXiv 

  Click for Model/Code and Paper
BenchCouncil's View on Benchmarking AI and Other Emerging Workloads

Dec 03, 2019
Jianfeng Zhan, Lei Wang, Wanling Gao, Rui Ren

This paper outlines BenchCouncil's view on the challenges, rules, and vision of benchmarking modern workloads like Big Data, AI or machine learning, and Internet Services. We conclude the challenges of benchmarking modern workloads as FIDSS (Fragmented, Isolated, Dynamic, Service-based, and Stochastic), and propose the PRDAERS benchmarking rules that the benchmarks should be specified in a paper-and-pencil manner, relevant, diverse, containing different levels of abstractions, specifying the evaluation metrics and methodology, repeatable, and scaleable. We believe proposing simple but elegant abstractions that help achieve both efficiency and general-purpose is the final target of benchmarking in future, which may be not pressing. In the light of this vision, we shortly discuss BenchCouncil's related projects.

* 7 pages 

  Click for Model/Code and Paper
Towards Amortized Ranking-Critical Training for Collaborative Filtering

Jun 10, 2019
Sam Lobel, Chunyuan Li, Jianfeng Gao, Lawrence Carin

Collaborative filtering is widely used in modern recommender systems. Recent research shows that variational autoencoders (VAEs) yield state-of-the-art performance by integrating flexible representations from deep neural networks into latent variable models, mitigating limitations of traditional linear factor models. VAEs are typically trained by maximizing the likelihood (MLE) of users interacting with ground-truth items. While simple and often effective, MLE-based training does not directly maximize the recommendation-quality metrics one typically cares about, such as top-N ranking. In this paper we investigate new methods for training collaborative filtering models based on actor-critic reinforcement learning, to directly optimize the non-differentiable quality metrics of interest. Specifically, we train a critic network to approximate ranking-based metrics, and then update the actor network (represented here by a VAE) to directly optimize against the learned metrics. In contrast to traditional learning-to-rank methods that require to re-run the optimization procedure for new lists, our critic-based method amortizes the scoring process with a neural network, and can directly provide the (approximate) ranking scores for new lists. Empirically, we show that the proposed methods outperform several state-of-the-art baselines, including recently-proposed deep learning approaches, on three large-scale real-world datasets. The code to reproduce the experimental results and figure plots is on Github:

* The first two authors contributed equally to this manuscript. Code: 

  Click for Model/Code and Paper
Budgeted Policy Learning for Task-Oriented Dialogue Systems

Jun 02, 2019
Zhirui Zhang, Xiujun Li, Jianfeng Gao, Enhong Chen

This paper presents a new approach that extends Deep Dyna-Q (DDQ) by incorporating a Budget-Conscious Scheduling (BCS) to best utilize a fixed, small amount of user interactions (budget) for learning task-oriented dialogue agents. BCS consists of (1) a Poisson-based global scheduler to allocate budget over different stages of training; (2) a controller to decide at each training step whether the agent is trained using real or simulated experiences; (3) a user goal sampling module to generate the experiences that are most effective for policy learning. Experiments on a movie-ticket booking task with simulated and real users show that our approach leads to significant improvements in success rate over the state-of-the-art baselines given the fixed budget.

* 10 pages, 7 figures, ACL 2019 

  Click for Model/Code and Paper
The Design and Implementation of XiaoIce, an Empathetic Social Chatbot

Dec 21, 2018
Li Zhou, Jianfeng Gao, Di Li, Heung-Yeung Shum

This paper describes the development of the Microsoft XiaoIce system, the most popular social chatbot in the world. XiaoIce is uniquely designed as an AI companion with an emotional connection to satisfy the human need for communication, affection, and social belonging. We take into account both intelligent quotient (IQ) and emotional quotient (EQ) in system design, cast human-machine social chat as decision-making over Markov Decision Processes (MDPs), and optimize XiaoIce for long-term user engagement, measured in expected Conversation-turns Per Session (CPS). We detail the system architecture and key components including dialogue manager, core chat, skills, and an empathetic computing module. We show how XiaoIce dynamically recognizes human feelings and states, understands user intents, and responds to user needs throughout long conversations. Since the release in 2014, XiaoIce has communicated with over 660 million users and succeeded in establishing long-term relationships with many of them. Analysis of large-scale online logs shows that XiaoIce has achieved an average CPS of 23, which is significantly higher than that of other chatbots and even human conversations.

  Click for Model/Code and Paper
Stochastic Answer Networks for Machine Reading Comprehension

May 15, 2018
Xiaodong Liu, Yelong Shen, Kevin Duh, Jianfeng Gao

We propose a simple yet robust stochastic answer network (SAN) that simulates multi-step reasoning in machine reading comprehension. Compared to previous work such as ReasoNet which used reinforcement learning to determine the number of steps, the unique feature is the use of a kind of stochastic prediction dropout on the answer module (final layer) of the neural network during the training. We show that this simple trick improves robustness and achieves results competitive to the state-of-the-art on the Stanford Question Answering Dataset (SQuAD), the Adversarial SQuAD, and the Microsoft MAchine Reading COmprehension Dataset (MS MARCO).

* 11 pages, 5 figures, Accepted to ACL 2018 

  Click for Model/Code and Paper
An Empirical Analysis of Multiple-Turn Reasoning Strategies in Reading Comprehension Tasks

Nov 09, 2017
Yelong Shen, Xiaodong Liu, Kevin Duh, Jianfeng Gao

Reading comprehension (RC) is a challenging task that requires synthesis of information across sentences and multiple turns of reasoning. Using a state-of-the-art RC model, we empirically investigate the performance of single-turn and multiple-turn reasoning on the SQuAD and MS MARCO datasets. The RC model is an end-to-end neural network with iterative attention, and uses reinforcement learning to dynamically control the number of turns. We find that multiple-turn reasoning outperforms single-turn reasoning for all question and answer types; further, we observe that enabling a flexible number of turns generally improves upon a fixed multiple-turn strategy. %across all question types, and is particularly beneficial to questions with lengthy, descriptive answers. We achieve results competitive to the state-of-the-art on these two datasets.

  Click for Model/Code and Paper
Bi-directional Attention with Agreement for Dependency Parsing

Sep 22, 2016
Hao Cheng, Hao Fang, Xiaodong He, Jianfeng Gao, Li Deng

We develop a novel bi-directional attention model for dependency parsing, which learns to agree on headword predictions from the forward and backward parsing directions. The parsing procedure for each direction is formulated as sequentially querying the memory component that stores continuous headword embeddings. The proposed parser makes use of {\it soft} headword embeddings, allowing the model to implicitly capture high-order parsing history without dramatically increasing the computational complexity. We conduct experiments on English, Chinese, and 12 other languages from the CoNLL 2006 shared task, showing that the proposed model achieves state-of-the-art unlabeled attachment scores on 6 languages.

* EMNLP 2016 

  Click for Model/Code and Paper
Learning Semantic Representations for the Phrase Translation Model

Nov 28, 2013
Jianfeng Gao, Xiaodong He, Wen-tau Yih, Li Deng

This paper presents a novel semantic-based phrase translation model. A pair of source and target phrases are projected into continuous-valued vector representations in a low-dimensional latent semantic space, where their translation score is computed by the distance between the pair in this new space. The projection is performed by a multi-layer neural network whose weights are learned on parallel training data. The learning is aimed to directly optimize the quality of end-to-end machine translation results. Experimental evaluation has been performed on two Europarl translation tasks, English-French and German-English. The results show that the new semantic-based phrase translation model significantly improves the performance of a state-of-the-art phrase-based statistical machine translation sys-tem, leading to a gain of 0.7-1.0 BLEU points.

  Click for Model/Code and Paper
Online Classification Using a Voted RDA Method

Oct 17, 2013
Tianbing Xu, Jianfeng Gao, Lin Xiao, Amelia Regan

We propose a voted dual averaging method for online classification problems with explicit regularization. This method employs the update rule of the regularized dual averaging (RDA) method, but only on the subsequence of training examples where a classification error is made. We derive a bound on the number of mistakes made by this method on the training set, as well as its generalization error rate. We also introduce the concept of relative strength of regularization, and show how it affects the mistake bound and generalization performance. We experimented with the method using $\ell_1$ regularization on a large-scale natural language processing task, and obtained state-of-the-art classification performance with fairly sparse models.

* 23 pages, 5 figures 

  Click for Model/Code and Paper
Multi-Task Learning for Machine Reading Comprehension

Sep 18, 2018
Yichong Xu, Xiaodong Liu, Yelong Shen, Jingjing Liu, Jianfeng Gao

We propose a multi-task learning framework to jointly train a Machine Reading Comprehension (MRC) model on multiple datasets across different domains. Key to the proposed method is to learn robust and general contextual representations with the help of out-domain data in a multi-task framework. Empirical study shows that the proposed approach is orthogonal to the existing pre-trained representation models, such as word embedding and language models. Experiments on the Stanford Question Answering Dataset (SQuAD), the Microsoft MAchine Reading COmprehension Dataset (MS MARCO), NewsQA and other datasets show that our multi-task learning approach achieves significant improvement over state-of-the-art models in most MRC tasks.

* 9 pages, 2 figures, 7 tables 

  Click for Model/Code and Paper
Navigating with Graph Representations for Fast and Scalable Decoding of Neural Language Models

Jun 11, 2018
Minjia Zhang, Xiaodong Liu, Wenhan Wang, Jianfeng Gao, Yuxiong He

Neural language models (NLMs) have recently gained a renewed interest by achieving state-of-the-art performance across many natural language processing (NLP) tasks. However, NLMs are very computationally demanding largely due to the computational cost of the softmax layer over a large vocabulary. We observe that, in decoding of many NLP tasks, only the probabilities of the top-K hypotheses need to be calculated preciously and K is often much smaller than the vocabulary size. This paper proposes a novel softmax layer approximation algorithm, called Fast Graph Decoder (FGD), which quickly identifies, for a given context, a set of K words that are most likely to occur according to a NLM. We demonstrate that FGD reduces the decoding time by an order of magnitude while attaining close to the full softmax baseline accuracy on neural machine translation and language modeling tasks. We also prove the theoretical guarantee on the softmax approximation quality.

  Click for Model/Code and Paper
Language-Based Image Editing with Recurrent Attentive Models

Jun 10, 2018
Jianbo Chen, Yelong Shen, Jianfeng Gao, Jingjing Liu, Xiaodong Liu

We investigate the problem of Language-Based Image Editing (LBIE). Given a source image and a natural language description, we want to generate a target image by editing the source image based on the description. We propose a generic modeling framework for two sub-tasks of LBIE: language-based image segmentation and image colorization. The framework uses recurrent attentive models to fuse image and language features. Instead of using a fixed step size, we introduce for each region of the image a termination gate to dynamically determine after each inference step whether to continue extrapolating additional information from the textual description. The effectiveness of the framework is validated on three datasets. First, we introduce a synthetic dataset, called CoSaL, to evaluate the end-to-end performance of our LBIE system. Second, we show that the framework leads to state-of-the-art performance on image segmentation on the ReferIt dataset. Third, we present the first language-based colorization result on the Oxford-102 Flowers dataset.

* Accepted to CVPR 2018 as a Spotlight 

  Click for Model/Code and Paper
Dynamic Fusion Networks for Machine Reading Comprehension

Feb 26, 2018
Yichong Xu, Jingjing Liu, Jianfeng Gao, Yelong Shen, Xiaodong Liu

This paper presents a novel neural model - Dynamic Fusion Network (DFN), for machine reading comprehension (MRC). DFNs differ from most state-of-the-art models in their use of a dynamic multi-strategy attention process, in which passages, questions and answer candidates are jointly fused into attention vectors, along with a dynamic multi-step reasoning module for generating answers. With the use of reinforcement learning, for each input sample that consists of a question, a passage and a list of candidate answers, an instance of DFN with a sample-specific network architecture can be dynamically constructed by determining what attention strategy to apply and how many reasoning steps to take. Experiments show that DFNs achieve the best result reported on RACE, a challenging MRC dataset that contains real human reading questions in a wide variety of types. A detailed empirical analysis also demonstrates that DFNs can produce attention vectors that summarize information from questions, passages and answer candidates more effectively than other popular MRC models.

* 13 pages, 5 figures, 5 tables 

  Click for Model/Code and Paper