Deep-learning models for language generation tasks tend to produce repetitive output. Various methods have been proposed to encourage lexical diversity during decoding, but this often comes at a cost to the perceived fluency and adequacy of the output. In this work, we propose to ameliorate this cost by using an Imitation Learning approach to explore the level of diversity that a language generation model can reliably produce. Specifically, we augment the decoding process with a meta-classifier trained to distinguish which words at any given timestep will lead to high-quality output. We focus our experiments on concept-to-text generation where models are sensitive to the inclusion of irrelevant words due to the strict relation between input and output. Our analysis shows that previous methods for diversity underperform in this setting, while human evaluation suggests that our proposed method achieves a high level of diversity with minimal effect to the output's fluency and adequacy.
Tel-Aviv University（特拉维夫大学）；Allen Institute for AI（艾伦人工智能研究所）
Despite growing interest in natural language generation (NLG) models that produce diverse outputs, there is currently no principled method for evaluating the diversity of an NLG system. In this work, we propose a framework for evaluating diversity metrics. The framework measures the correlation between a proposed diversity metric and a diversity parameter, a single parameter that controls some aspect of diversity in generated text. For example, a diversity parameter might be a binary variable used to instruct crowdsourcing workers to generate text with either low or high content diversity. We …
Mila（米拉）；Université de Montréal（蒙特利尔大学）；McGill University（麦吉尔大学）；Google Brain（谷歌大脑）；Facebook AI Research（脸书人工智能研究院）；HEC Montréal（蒙特利尔高等商学院）
Generating high-quality text with sufficient diversity is essential for a wide range of Natural Language Generation (NLG) tasks. Maximum-Likelihood (MLE) models trained with teacher forcing have consistently been reported as weak baselines, where poor performance is attributed to exposure bias (Bengio et al., 2015; Ranzato et al., 2015); at inference time, the model is fed its own prediction instead of a ground-truth token, which can lead to accumulating errors and poor samples. This line of reasoning has led to an outbreak of adversarial based approaches for NLG, on the account that GANs do n…
Code-mixing is a phenomenon of mixing words and phrases from two or more languages in a single utterance of speech and text. Due to the high linguistic diversity, code-mixing presents several challenges in evaluating standard natural language generation (NLG) tasks. Various widely popular metrics perform poorly with the code-mixed NLG tasks. To address this challenge, we present a metric independent evaluation pipeline MIPE that significantly improves the correlation between evaluation metrics and human judgments on the generated code-mixed text. As a use case, we demonstrate the performance o…
Transformer-based language models have shown to be very powerful for natural language generation (NLG). However, text generation conditioned on some user inputs, such as topics or attributes, is non-trivial. Past approach relies on either modifying the original LM architecture, re-training the LM on corpora with attribute labels, or having separately trained `guidance models' to guide text generation in decoding. We argued that the above approaches are not necessary, and the original unconditioned LM is sufficient for conditioned NLG. We evaluated our approaches by the samples' fluency and div…
Evaluating Evaluation Metrics: A Framework for Analyzing NLG Evaluation Metrics using Measurement Theory
Visa Research（Visa研究院）；Microsoft Research Montréal（微软蒙特利尔研究院）；Johns Hopkins University（约翰斯·霍普金斯大学）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
This paper provides a comprehensive analysis of the first shared task on End-to-End Natural Language Generation (NLG) and identifies avenues for future research based on the results. This shared task aimed to assess whether recent end-to-end NLG systems can generate more complex output by learning from datasets containing higher lexical richness, syntactic complexity and diverse discourse phenomena. Introducing novel automatic and human metrics, we compare 62 systems submitted by 17 institutions, covering a wide range of approaches, including machine learning architectures -- with the majority…
Score function-based natural language generation (NLG) approaches such as REINFORCE, in general, suffer from low sample efficiency and training instability problems. This is mainly due to the non-differentiable nature of the discrete space sampling and thus these methods have to treat the discriminator as a black box and ignore the gradient information. To improve the sample efficiency and reduce the variance of REINFORCE, we propose a novel approach, TaylorGAN, which augments the gradient estimation by off-policy update and the first-order Taylor expansion. This approach enables us to train N…
Neural network based approaches to data-to-text natural language generation (NLG) have gained popularity in recent years, with the goal of generating a natural language prompt that accurately realizes an input meaning representation. To facilitate the training of neural network models, researchers created large datasets of paired utterances and their meaning representations. However, the creation of such datasets is an arduous task and they mostly consist of simple meaning representations composed of slot and value tokens to be realized. These representations do not include any contextual info…
Evaluation is a bottleneck in the development of natural language generation (NLG) models. Automatic metrics such as BLEU rely on references, but for tasks such as open-ended generation, there are no references to draw upon. Although language diversity can be estimated using statistical measures such as perplexity, measuring language quality requires human evaluation. However, because human evaluation at scale is slow and expensive, it is used sparingly; it cannot be used to rapidly iterate on NLG models, in the way BLEU is used for machine translation. To this end, we propose BLEU Neighbors, …
We present DeepGen, a system deployed at web scale for automatically creating sponsored search advertisements (ads) for BingAds customers. We leverage state-of-the-art natural language generation (NLG) models to generate fluent ads from advertiser's web pages in an abstractive fashion and solve practical issues such as factuality and inference speed. In addition, our system creates a customized ad in real-time in response to the user's search query, therefore highlighting different aspects of the same product based on what the user is looking for. To achieve this, our system generates a divers…
The uptake of deep learning in natural language generation (NLG) led to the release of both small and relatively large parallel corpora for training neural models. The existing data-to-text datasets are, however, aimed at task-oriented dialogue systems, and often thus limited in diversity and versatility. They are typically crowdsourced, with much of the noise left in them. Moreover, current neural NLG models do not take full advantage of large training data, and due to their strong generalizing properties produce sentences that look template-like regardless. We therefore present a new corpus …
Modern natural language generation paradigms require a good decoding strategy to obtain quality sequences out of the model. Beam search yields high-quality but low diversity outputs; stochastic approaches suffer from high variance and sometimes low quality, but the outputs tend to be more natural and creative. In this work, we propose a deterministic search algorithm balancing both quality and diversity. We first investigate the vanilla best-first search (BFS) algorithm and then propose the Best-$k$ Search algorithm. Inspired by BFS, we greedily expand the top $k$ nodes, instead of only the fi…
University of Virginia（弗吉尼亚大学）；The Chinese University of Hong Kong（香港中文大学）
Generating high quality texts with high diversity is important for many NLG applications, but current methods mostly focus on building deterministic models to generate higher quality texts and do not provide many options for promoting diversity. In this work, we present a novel latent structured variable model to generate high quality texts by enriching contextual representation learning of encoder-decoder models. Specifically, we introduce a stochastic function to map deterministic encoder hidden states into random context variables. The proposed stochastic function is sampled from a Gaussian…
Comment generation, a new and challenging task in Natural Language Generation (NLG), attracts a lot of attention in recent years. However, comments generated by previous work tend to lack pertinence and diversity. In this paper, we propose a novel generation model based on Topic-aware Pointer-Generator Networks (TPGN), which can utilize the topic information hidden in the articles to guide the generation of pertinent and diversified comments. Firstly, we design a keyword-level and topic-level encoder attention mechanism to capture topic information in the articles. Next, we integrate the topic…
Training generative models that can generate high-quality text with sufficient diversity is an important open problem for Natural Language Generation (NLG) community. Recently, generative adversarial models have been applied extensively on text generation tasks, where the adversarially trained generators alleviate the exposure bias experienced by conventional maximum likelihood approaches and result in promising generation quality. However, due to the notorious defect of mode collapse for adversarial training, the adversarially trained generators face a quality-diversity trade-off, i.e., the g…
University of Amsterdam（阿姆斯特丹大学）；Max Planck Institute for Human Development（马克斯·普朗克人类发展研究所）
The release of openly available, robust natural language generation algorithms (NLG) has spurred much public attention and debate. One reason lies in the algorithms' purported ability to generate human-like text across various domains. Empirical evidence using incentivized tasks to assess whether people (a) can distinguish and (b) prefer algorithm-generated versus human-written text is lacking. We conducted two experiments assessing behavioral reactions to the state-of-the-art Natural Language Generation algorithm GPT-2 (Ntotal = 830). Using the identical starting lines of human poems, GPT-2 p…
Alibaba Group（阿里巴巴集团）；Wuhan University（武汉大学）；University of Michigan（密歇根大学）；Amazon（亚马逊）；Tencent AI Lab（腾讯人工智能实验室）
Conditional Text Generation has drawn much attention as a topic of Natural Language Generation (NLG) which provides the possibility for humans to control the properties of generated contents. Current conditional generation models cannot handle emerging conditions due to their joint end-to-end learning fashion. When a new condition added, these techniques require full retraining. In this paper, we present a new framework named Pre-train and Plug-in Variational Auto-Encoder (PPVAE) towards flexible conditional text generation. PPVAE decouples the text generation module from the condition represe…
We introduce CGA, a conditional VAE architecture, to control, generate, and augment text. CGA is able to generate natural English sentences controlling multiple semantic and syntactic attributes by combining adversarial learning with a context-aware loss and a cyclical word dropout routine. We demonstrate the value of the individual model components in an ablation study. The scalability of our approach is ensured through a single discriminator, independently of the number of attributes. We show high quality, diversity and attribute control in the generated sentences through a series of automat…
Cross-domain natural language generation (NLG) is still a difficult task within spoken dialogue modelling. Given a semantic representation provided by the dialogue manager, the language generator should generate sentences that convey desired information. Traditional template-based generators can produce sentences with all necessary information, but these sentences are not sufficiently diverse. With RNN-based models, the diversity of the generated sentences can be high, however, in the process some information is lost. In this work, we improve an RNN-based generator by considering latent inform…
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Countermeasures to effectively fight the ever increasing hate speech online without blocking freedom of speech is of great social interest. Natural Language Generation (NLG), is uniquely capable of developing scalable solutions. However, off-the-shelf NLG methods are primarily sequence-to-sequence neural models and they are limited in that they generate commonplace, repetitive and safe responses regardless of the hate speech (e.g., "Please refrain from using such language.") or irrelevant responses, making them ineffective for de-escalating hateful conversations. In this paper, we design a thr…
Allen Institute for AI（艾伦人工智能研究所）；Amazon Alexa AI（亚马逊Alexa人工智能部门）；Bangladesh University of Engineering and Technology（孟加拉国工程技术大学）；Carnegie Mellon University（卡内基梅隆大学）；Charles University（查理大学）
Evaluation in machine learning is usually informed by past choices, for example which datasets or metrics to use. This standardization enables the comparison on equal footing using leaderboards, but the evaluation choices become sub-optimal as better alternatives arise. This problem is especially pertinent in natural language generation which requires ever-improving suites of datasets, metrics, and human evaluation to make definitive claims. To make following best model evaluation practices easier, we introduce GEMv2. The new version of the Generation, Evaluation, and Metrics Benchmark introdu…
From Random to Informed Data Selection: A Diversity-Based Approach to Optimize Human Annotation and Few-Shot Learning
Escola Politécnica, Universidade de São Paulo（圣保罗大学理工学院）；Télécom Paris, Institut Polytechnique de Paris（巴黎高等电信学校，巴黎综合理工学院）；Faculty of Economics, University of Porto（波尔图大学经济学院）；INESC TEC（葡萄牙系统工程与计算机技术研究所）
Beyond Hearing: Learning Task-Agnostic ExG Representations from Earphones via Physiology-Informed Tokenization
KAIST（韩国科学技术院）；Carnegie Mellon University（卡内基梅隆大学）；University of Cambridge（剑桥大学）；Shanghai Jiao Tong University（上海交通大学）；Pennsylvania State University（宾夕法尼亚州立大学）；UCLA（加州大学洛杉矶分校）；Northwest University（北华大学）；University of Texas at Austin（德克萨斯大学奥斯汀分校）；Microsoft Research（微软研究院）
Department of Computer Science, Cornell University（康奈尔大学计算机科学系）；Statistics and Data Science Center, Massachusetts Institute of Technology（麻省理工学院统计与数据科学中心）；Department of Statistics, Cornell University（康奈尔大学统计学系）
Sparks of Science: Hypothesis Generation Using Structured Paper Data
University of Oxford（牛津大学）；Oak Ridge National Laboratory（橡树岭国家实验室）；University College London（伦敦大学学院）；University of Toronto（多伦多大学）；Australian National University（澳大利亚国立大学）；Modulos AG（Modulos AG公司）；Stanford University（斯坦福大学）
