GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection

Pre-trained models have proved to be powerful in enhancing task-oriented dialog systems. However, current pre-training methods mainly focus on enhancing dialog understanding and generation tasks while neglecting the exploitation of dialog policy. In this paper, we propose GALAXY, a novel pre-trained dialog model that explicitly learns dialog policy from limited labeled dialogs and large-scale unlabeled dialog corpora via semi-supervised learning. Specifically, we introduce a dialog act prediction task for policy optimization during pre-training and employ a consistency regularization term to refine the learned representation with the help of unlabeled dialogs. We also implement a gating mechanism to weigh suitable unlabeled dialog samples. Empirical results show that GALAXY substantially improves the performance of task-oriented dialog systems, and achieves new state-of-the-art results on benchmark datasets: In-Car, MultiWOZ2.0 and MultiWOZ2.1, improving their end-to-end combined scores by 2.5, 5.3 and 5.5 points, respectively. We also show that GALAXY has a stronger few-shot ability than existing models under various low-resource settings.

Knowledge Based Multilingual Language Model

Knowledge enriched language representation learning has shown promising performance across various knowledge-intensive NLP tasks. However, existing knowledge based language models are all trained with monolingual knowledge graph data, which limits their application to more languages. In this work, we present a novel framework to pretrain knowledge based multilingual language models (KMLMs). We first generate a large amount of code-switched synthetic sentences and reasoning-based multilingual training data using the Wikidata knowledge graphs. Then based on the intra- and inter-sentence structures of the generated data, we design pretraining tasks to facilitate knowledge learning, which allows the language models to not only memorize the factual knowledge but also learn useful logical patterns. Our pretrained KMLMs demonstrate significant performance improvements on a wide range of knowledge-intensive cross-lingual NLP tasks, including named entity recognition, factual knowledge retrieval, relation classification, and a new task designed by us, namely, logic reasoning. Our code and pretrained language models will be made publicly available.

Achieving Human Parity on Visual Question Answering

The Visual Question Answering (VQA) task utilizes both visual image and language analysis to answer a textual question with respect to an image. It has been a popular research topic with an increasing number of real-world applications in the last decade. This paper describes our recent research of AliceMind-MMU (ALIbaba's Collection of Encoder-decoders from Machine IntelligeNce lab of Damo academy - MultiMedia Understanding) that obtains similar or even slightly better results than human being does on VQA. This is achieved by systematically improving the VQA pipeline including: (1) pre-training with comprehensive visual and textual feature representation; (2) effective cross-modal interaction with learning to attend; and (3) A novel knowledge mining framework with specialized expert modules for the complex VQA task. Treating different types of visual questions with corresponding expertise needed plays an important role in boosting the performance of our VQA architecture up to the human level. An extensive set of experiments and analysis are conducted to demonstrate the effectiveness of the new research work.

GlobalWoZ: Globalizing MultiWoZ to Develop Multilingual Task-Oriented Dialogue Systems

Much recent progress in task-oriented dialogue (ToD) systems has been driven by available annotation data across multiple domains for training. Over the last few years, there has been a move towards data curation for multilingual ToD systems that are applicable to serve people speaking different languages. However, existing multilingual ToD datasets either have a limited coverage of languages due to the high cost of data curation, or ignore the fact that dialogue entities barely exist in countries speaking these languages. To tackle these limitations, we introduce a novel data curation method that generates GlobalWoZ -- a large-scale multilingual ToD dataset globalized from an English ToD dataset for three unexplored use cases. Our method is based on translating dialogue templates and filling them with local entities in the target-language countries. We release our dataset as well as a set of strong baselines to encourage research on learning multilingual ToD systems for real use cases.

MReD: A Meta-Review Dataset for Controllable Text Generation

When directly using existing text generation datasets for controllable generation, we are facing the problem of not having the domain knowledge and thus the aspects that could be controlled are limited.A typical example is when using CNN/Daily Mail dataset for controllable text summarization, there is no guided information on the emphasis of summary sentences. A more useful text generator should leverage both the input text and control variables to guide the generation, which can only be built with deep understanding of the domain knowledge. Motivated by this vi-sion, our paper introduces a new text generation dataset, named MReD. Our new dataset consists of 7,089 meta-reviews and all its 45k meta-review sentences are manually annotated as one of the carefully defined 9 categories, including abstract, strength, decision, etc. We present experimental results on start-of-the-art summarization models, and propose methods for controlled generation on both extractive and abstractive models using our annotated data. By exploring various settings and analaysing the model behavior with respect to the control inputs, we demonstrate the challenges and values of our dataset. MReD allows us to have a better understanding of the meta-review corpora and enlarge the research room for controllable text generation.

DialogueCSE: Dialogue-based Contrastive Learning of Sentence Embeddings

Learning sentence embeddings from dialogues has drawn increasing attention due to its low annotation cost and high domain adaptability. Conventional approaches employ the siamese-network for this task, which obtains the sentence embeddings through modeling the context-response semantic relevance by applying a feed-forward network on top of the sentence encoders. However, as the semantic textual similarity is commonly measured through the element-wise distance metrics (e.g. cosine and L2 distance), such architecture yields a large gap between training and evaluating. In this paper, we propose DialogueCSE, a dialogue-based contrastive learning approach to tackle this issue. DialogueCSE first introduces a novel matching-guided embedding (MGE) mechanism, which generates a context-aware embedding for each candidate response embedding (i.e. the context-free embedding) according to the guidance of the multi-turn context-response matching matrices. Then it pairs each context-aware embedding with its corresponding context-free embedding and finally minimizes the contrastive loss across all pairs. We evaluate our model on three multi-turn dialogue datasets: the Microsoft Dialogue Corpus, the Jing Dong Dialogue Corpus, and the E-commerce Dialogue Corpus. Evaluation results show that our approach significantly outperforms the baselines across all three datasets in terms of MAP and Spearman's correlation measures, demonstrating its effectiveness. Further quantitative experiments show that our approach achieves better performance when leveraging more dialogue context and remains robust when less training data is provided.

LightNER: A Lightweight Generative Framework with Prompt-guided Attention for Low-resource NER

Most existing NER methods rely on extensive labeled data for model training, which struggles in the low-resource scenarios with limited training data. Recently, prompt-tuning methods for pre-trained language models have achieved remarkable performance in few-shot learning by exploiting prompts as task guidance to reduce the gap between training progress and downstream tuning. Inspired by prompt learning, we propose a novel lightweight generative framework with prompt-guided attention for low-resource NER (LightNER). Specifically, we construct the semantic-aware answer space of entity categories for prompt learning to generate the entity span sequence and entity categories without any label-specific classifiers. We further propose prompt-guided attention by incorporating continuous prompts into the self-attention layer to re-modulate the attention and adapt pre-trained weights. Note that we only tune those continuous prompts with the whole parameter of the pre-trained language model fixed, thus, making our approach lightweight and flexible for low-resource scenarios and can better transfer knowledge across domains. Experimental results show that LightNER can obtain comparable performance in the standard supervised setting and outperform strong baselines in low-resource settings by tuning only a small part of the parameters.

MELM: Data Augmentation with Masked Entity Language Modeling for Cross-lingual NER

Data augmentation for cross-lingual NER requires fine-grained control over token labels of the augmented text. Existing augmentation approach based on masked language modeling may replace a labeled entity with words of a different class, which makes the augmented sentence incompatible with the original label sequence, and thus hurts the performance.We propose a data augmentation framework with Masked-Entity Language Modeling (MELM) which effectively ensures the replacing entities fit the original labels. Specifically, MELM linearizes NER labels into sentence context, and thus the fine-tuned MELM is able to predict masked tokens by explicitly conditioning on their labels. Our MELM is agnostic to the source of data to be augmented. Specifically, when MELM is applied to augment training data of the source language, it achieves up to 3.5% F1 score improvement for cross-lingual NER. When unlabeled target data is available and MELM can be further applied to augment pseudo-labeled target data, the performance gain reaches 5.7%. Moreover, MELM consistently outperforms multiple baseline methods for data augmentation.

CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark

Artificial Intelligence (AI), along with the recent progress in biomedical language understanding, is gradually changing medical practice. With the development of biomedical language understanding benchmarks, AI applications are widely used in the medical field. However, most benchmarks are limited to English, which makes it challenging to replicate many of the successes in English for other languages. To facilitate research in this direction, we collect real-world biomedical data and present the first Chinese Biomedical Language Understanding Evaluation (CBLUE) benchmark: a collection of natural language understanding tasks including named entity recognition, information extraction, clinical diagnosis normalization, single-sentence/sentence-pair classification, and an associated online platform for model evaluation, comparison, and analysis. To establish evaluation on these tasks, we report empirical results with the current 11 pre-trained Chinese models, and experimental results show that state-of-the-art neural models perform by far worse than the human ceiling. Our benchmark is released at \url{https://tianchi.aliyun.com/dataset/dataDetail?dataId=95414&lang=en-us}.