Journal Article•
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
TL;DR: This article introduced a unified framework that converts all text-based language problems into a text-to-text format and compared pre-training objectives, architectures, unlabeled data sets, transfer approaches, and other factors on dozens of language understanding tasks.
Abstract: Transfer learning, where a model is first pre-trained on a data-rich task before being fine-tuned on a downstream task, has emerged as a powerful technique in natural language processing (NLP). The effectiveness of transfer learning has given rise to a diversity of approaches, methodology, and practice. In this paper, we explore the landscape of transfer learning techniques for NLP by introducing a unified framework that converts all text-based language problems into a text-to-text format. Our systematic study compares pre-training objectives, architectures, unlabeled data sets, transfer approaches, and other factors on dozens of language understanding tasks. By combining the insights from our exploration with scale and our new ``Colossal Clean Crawled Corpus'', we achieve state-of-the-art results on many benchmarks covering summarization, question answering, text classification, and more. To facilitate future work on transfer learning for NLP, we release our data set, pre-trained models, and code.
Citations
More filters
Proceedings Article•
01 Nov 2021TL;DR: Li et al. as mentioned in this paper proposed to train individual dense passage retrievers for different tasks and aggregate their predictions during test time, where they use uncertainty estimation as weights to indicate how probable a specific query belongs to each expert's expertise.
Abstract: Multi-task dense retrieval models can be used to retrieve documents from a common corpus (e.g., Wikipedia) for different open-domain question-answering (QA) tasks. However, (CITATION) shows that jointly learning different QA tasks with one dense model is not always beneficial due to corpus inconsistency. For example, SQuAD only focuses on a small set of Wikipedia articles while datasets like NQ and Trivia cover more entries, and joint training on their union can cause performance degradation. To solve this problem, we propose to train individual dense passage retrievers (DPR) for different tasks and aggregate their predictions during test time, where we use uncertainty estimation as weights to indicate how probable a specific query belongs to each expert’s expertise. Our method reaches state-of-the-art performance on 5 benchmark QA datasets, with up to 10% improvement in top-100 accuracy compared to a joint-training multi-task DPR on SQuAD. We also show that our method handles corpus inconsistency better than the joint-training DPR on a mixed subset of different QA datasets. Code and data are available at https://github.com/alexlimh/DPR_MUF.
3 citations
TL;DR: This work presents a pretraining financial text encoder, named F-BERT, a domain-specific language model pretrained on large-scale financial corpora, which achieves strong results on several financial text mining tasks.
Abstract: As the number of financial literature grows rapidly, Financial text mining is becoming important increasingly. In recent years, extracting valuable information from financial documents, namely financial text mining, gained significant popularity within research communities. Although Deep Learning-based financial text mining has achieved remarkable progress recently, in financial fields it still suffers from issues of lack of task-specific labeled training data. To alleviate these issues, we present a pretraining financial text encoder, named F-BERT, a domain-specific language model pretrained on large-scale financial corpora. Different from original BERT, proposed F-BERT is trained continually on both general corpus and financial domain corpus, and four pretraining tasks can be pretrained through lifelong learning, which can enable our F-BERT to continually capture language knowledge and semantic information. The experimental results demonstrate that proposed F-BERT achieves strong results on several financial text mining tasks. Extensive experimental results show the effectiveness and robustness of F-BERT. The source code and pretrained models of F-BERT are available online.
3 citations
Posted Content•
TL;DR: ExplaGraphs as mentioned in this paper is a commonsense-augmented graph collection dataset for stance prediction, where a model has to predict whether the argument supports or counters the belief and also generate a non-trivial, complete, and unambiguous explanation for the predicted stance.
Abstract: Recent commonsense-reasoning tasks are typically discriminative in nature, where a model answers a multiple-choice question for a certain context. Discriminative tasks are limiting because they fail to adequately evaluate the model's ability to reason and explain predictions with underlying commonsense knowledge. They also allow such models to use reasoning shortcuts and not be "right for the right reasons". In this work, we present ExplaGraphs, a new generative and structured commonsense-reasoning task (and an associated dataset) of explanation graph generation for stance prediction. Specifically, given a belief and an argument, a model has to predict whether the argument supports or counters the belief and also generate a commonsense-augmented graph that serves as non-trivial, complete, and unambiguous explanation for the predicted stance. The explanation graphs for our dataset are collected via crowdsourcing through a novel Collect-Judge-And-Refine graph collection framework that improves the graph quality via multiple rounds of verification and refinement. A significant 83% of our graphs contain external commonsense nodes with diverse structures and reasoning depths. We also propose a multi-level evaluation framework that checks for the structural and semantic correctness of the generated graphs and their plausibility with human-written graphs. We experiment with state-of-the-art text generation models like BART and T5 to generate explanation graphs and observe that there is a large gap with human performance, thereby encouraging useful future work for this new commonsense graph-based explanation generation task.
3 citations
Posted Content•
TL;DR: In this article, a large-scale Transformer-based pre-training dataset with over a billion images was generated via weakly-supervised pretraining to improve the performance of these visual representations.
Abstract: Large-scale pretraining of visual representations has led to state-of-the-art performance on a range of benchmark computer vision tasks, yet the benefits of these techniques at extreme scale in complex production systems has been relatively unexplored. We consider the case of a popular visual discovery product, where these representations are trained with multi-task learning, from use-case specific visual understanding (e.g. skin tone classification) to general representation learning for all visual content (e.g. embeddings for retrieval). In this work, we describe how we (1) generate a dataset with over a billion images via large weakly-supervised pretraining to improve the performance of these visual representations, and (2) leverage Transformers to replace the traditional convolutional backbone, with insights into both system and performance improvements, especially at 1B+ image scale. To support this backbone model, we detail a systematic approach to deriving weakly-supervised image annotations from heterogenous text signals, demonstrating the benefits of clustering techniques to handle the long-tail distribution of image labels. Through a comprehensive study of offline and online evaluation, we show that large-scale Transformer-based pretraining provides significant benefits to industry computer vision applications. The model is deployed in a production visual shopping system, with 36% improvement in top-1 relevance and 23% improvement in click-through volume. We conduct extensive experiments to better understand the empirical relationships between Transformer-based architectures, dataset scale, and the performance of production vision systems.
3 citations
Posted Content•
TL;DR: Zhang et al. as discussed by the authors formalized text-to-table information extraction as a sequence to sequence (seq2seq) problem, and employed a seq2seq model fine-tuned from a pre-trained language model to perform the task.
Abstract: We study a new problem setting of information extraction (IE), referred to as text-to-table, which can be viewed as an inverse problem of the well-studied table-to-text. In text-to-table, given a text, one creates a table or several tables expressing the main content of the text, while the model is learned from text-table pair data. The problem setting differs from those of the existing methods for IE. First, the extraction can be carried out from long texts to large tables with complex structures. Second, the extraction is entirely data-driven, and there is no need to explicitly define the schemas. As far as we know, there has been no previous work that studies the problem. In this work, we formalize text-to-table as a sequence-to-sequence (seq2seq) problem. We first employ a seq2seq model fine-tuned from a pre-trained language model to perform the task. We also develop a new method within the seq2seq approach, exploiting two additional techniques in table generation: table constraint and table relation embeddings. We make use of four existing table-to-text datasets in our experiments on text-to-table. Experimental results show that the vanilla seq2seq model can outperform the baseline methods of using relation extraction and named entity extraction. The results also show that our method can further boost the performances of the vanilla seq2seq model. We further discuss the main challenges of the proposed task. The code and data will be made publicly available.
3 citations