Cross-lingual document representations enable language understanding in multilingual contexts and allow transfer learning from high-resource to low-resource languages at the document level. Recently large pre-trained language models such as BERT, XLM and XLM-RoBERTa have achieved great success when fine-tuned on sentence-level downstream tasks. It is tempting to apply these cross-lingual models to document representation learning. However, there are two challenges: (1) these models impose high costs on long document processing and thus many of them have strict length limit; (2) model fine-tuning requires extra data and computational resources, which is not practical in resource-limited settings. In this work, we address these challenges by proposing unsupervised Language-Agnostic Weighted Document Representations (LAWDR). We study the geometry of pre-trained sentence embeddings and leverage it to derive document representations without fine-tuning. Evaluated on cross-lingual document alignment, LAWDR demonstrates comparable performance to state-of-the-art models on benchmark datasets.
ABCD-LINK: Annotation Bootstrapping for Cross-Document Fine-Grained Links
Ubiquitous Knowledge Processing Lab (UKP Lab), Department of Computer Science and Hessian Center for AI (hessian.AI), TU Darmstadt（图腾知识处理实验室（UKP实验室）、计算机科学系和海德堡人工智能中心（hessian.AI）、图腾达姆斯塔特大学）；Hebrew University of Jerusalem and The Allen Institute for AI (AI2)（耶路撒冷希伯来大学和人工智能研究所（AI2））
Structured Legal Document Generation in India: A Model-Agnostic Wrapper Approach with VidhikDastaavej
Indian Institute of Technology Kanpur, India（印度理工学院坎浦分校）；University of Birmingham Dubai, UAE（迪拜大学）；Symbiosis Law School Pune, India（萨比欧法学院皮尔分校）；IISER Kolkata, India（印度科学学院科钦分校）
Two Six Labs, LLC.（Two Six Labs有限责任公司）；Sophos（Sophos公司）
Code search and comprehension have become more difficult in recent years due to the rapid expansion of available source code. Current tools lack a way to label arbitrary code at scale while maintaining up-to-date representations of new programming languages, libraries, and functionalities. Comprehensive labeling of source code enables users to search for documents of interest and obtain a high-level understanding of their contents. We use Stack Overflow code snippets and their tags to train a language-agnostic, deep convolutional neural network to automatically predict semantic labels for sour…
Bosch Center for Artificial Intelligence, Robert Bosch GmbH（罗伯特·博世有限公司博世人工智能中心）；Saarland University（萨尔兰大学）
The field of natural language processing (NLP) has recently seen a large change towards using pre-trained language models for solving almost any task. Despite showing great improvements in benchmark datasets for various tasks, these models often perform sub-optimal in non-standard domains like the clinical domain where a large gap between pre-training documents and target documents is observed. In this paper, we aim at closing this gap with domain-specific training of the language model and we investigate its effect on a diverse set of downstream tasks and settings. We introduce the pre-traine…
Architecture-Agnostic Curriculum Learning for Document Understanding: Empirical Evidence from Text-Only and Multimodal
1 Synchromedia Laboratory, \' E cole de Technologie Sup\' e rieure (\' E TS), 1100 Notre-Dame St W, Montreal, QC H3C 1K3, Canada；2 Department of Computer Science, University of Bari Aldo Moro, Via Orabona 4, 70125 Bari, Italy
The state of the art of handling rich morphology in neural machine translation (NMT) is to break word forms into subword units, so that the overall vocabulary size of these units fits the practical limits given by the NMT model and GPU memory capacity. In this paper, we compare two common but linguistically uninformed methods of subword construction (BPE and STE, the method implemented in Tensor2Tensor toolkit) and two linguistically-motivated methods: Morfessor and one novel method, based on a derivational dictionary. Our experiments with German-to-Czech translation, both morphologically rich…
We propose SelfDoc, a task-agnostic pre-training framework for document image understanding. Because documents are multimodal and are intended for sequential reading, our framework exploits the positional, textual, and visual information of every semantically meaningful component in a document, and it models the contextualization between each block of content. Unlike existing document pre-training models, our model is coarse-grained instead of treating individual words as input, therefore avoiding an overly fine-grained with excessive contextualization. Beyond that, we introduce cross-modal le…
University of Chinese Academy of Sciences（中国科学院大学）；Meituan（美团）
Recently, the retrieval models based on dense representations have been gradually applied in the first stage of the document retrieval tasks, showing better performance than traditional sparse vector space models. To obtain high efficiency, the basic structure of these models is Bi-encoder in most cases. However, this simple structure may cause serious information loss during the encoding of documents since the queries are agnostic. To address this problem, we design a method to mimic the queries on each of the documents by an iterative clustering process and represent the documents by multipl…
The Entity Disambiguation and Linking (EDL) task matches entity mentions in text to a unique Knowledge Base (KB) identifier such as a Wikipedia or Freebase id. It plays a critical role in the construction of a high quality information network, and can be further leveraged for a variety of information retrieval and NLP tasks such as text categorization and document tagging. EDL is a complex and challenging problem due to ambiguity of the mentions and real world text being multi-lingual. Moreover, EDL systems need to have high throughput and should be lightweight in order to scale to large datas…
J.P. Morgan AI Research（摩根大通人工智能研究（J.P. Morgan AI Research））
Task specific fine-tuning of a pre-trained neural language model using a custom softmax output layer is the de facto approach of late when dealing with document classification problems. This technique is not adequate when labeled examples are not available at training time and when the metadata artifacts in a document must be exploited. We address these challenges by generating document representations that capture both text and metadata artifacts in a task agnostic manner. Instead of traditional auto-regressive or auto-encoding based training, our novel self-supervised approach learns a soft-…
Jožef Stefan Institute（约热夫·斯泰凡研究所）；Jožef Stefan International Postgraduate School（约热夫·斯泰凡国际研究生学院）
Contemporary natural language processing (NLP) revolves around learning from latent document representations, generated either implicitly by neural language models or explicitly by methods such as doc2vec or similar. One of the key properties of the obtained representations is their dimension. Whilst the commonly adopted dimensions of 256 and 768 offer sufficient performance on many tasks, it is many times unclear whether the default dimension is the most suitable choice for the subsequent downstream learning tasks. Furthermore, representation dimensions are seldom subject to hyperparameter tu…
