This technical report briefly describes our JDExplore d-team's Vega v2 submission on the SuperGLUE leaderboard. SuperGLUE is more challenging than the widely used general language understanding evaluation (GLUE) benchmark, containing eight difficult language understanding tasks, including question answering, natural language inference, word sense disambiguation, coreference resolution, and reasoning. [Method] Instead of arbitrarily increasing the size of a pretrained language model (PLM), our aim is to 1) fully extract knowledge from the input pretraining data given a certain parameter budget, e.g., 6B, and 2) effectively transfer this knowledge to downstream tasks. To achieve goal 1), we propose self-evolution learning for PLMs to wisely predict the informative tokens that should be masked, and supervise the masked language modeling (MLM) process with rectified smooth labels. For goal 2), we leverage the prompt transfer technique to improve the low-resource tasks by transferring the knowledge from the foundation model and related downstream tasks to the target task. [Results] According to our submission record (Oct. 2022), with our optimized pretraining and fine-tuning strategies, our 6B Vega method achieved new state-of-the-art performance on 4/8 tasks, sitting atop the SuperGLUE leaderboard on Oct. 8, 2022, with an average score of 91.3.

迅速调整，它冻结了预审计的语言模型（PLM），只有微调的几个额外软提示的参数，在PLM具有数十亿个参数时，对全参数微调（即模型调整）显示出具有竞争性的性能，但仍然显示出竞争力。在较小的PLM的情况下，性能差。因此，迅速转移（POT），通过训练有素的类似源任务的提示来初始化目标提示，最近提议改善及时调整。但是，这样的香草锅方法通常会实现次优的性能，因为（i）锅对源目标对的相似性和（ii）直接对目标提示进行初始提示的提示敏感，而目标任务可能会导致灾难性忘记来源知识。为了解决这些问题，我们提出了一个新的指标，以准确预测及时的转移性（关于（i）），以及一种利用知识蒸馏技术将“知识”从源提示转移到的新颖的锅方法（即熊猫）目标以微妙的方式提示，并有效缓解灾难性遗忘（关于（ii））。此外，为了实现每个源目标对的自适应及时转移，我们使用指标来控制熊猫方法中的知识转移。对PLM的5个量表的21个源和9个目标数据集的189组组合进行了广泛而系统的实验，表明：1）我们提出的指标很好地预测了及时的可传递性； 2）在所有任务和型号中，我们的熊猫始终优于香草锅的平均得分2.3％（最高24.1％）； 3）通过我们的熊猫方法，及时调整可以比在各种PLM量表场景中的模型调整来实现竞争性甚至更好的性能。接受代码和模型将在接受后发布。

Language model pre-training, such as BERT, has significantly improved the performances of many natural language processing tasks. However, pre-trained language models are usually computationally expensive, so it is difficult to efficiently execute them on resourcerestricted devices. To accelerate inference and reduce model size while maintaining accuracy, we first propose a novel Transformer distillation method that is specially designed for knowledge distillation (KD) of the Transformer-based models. By leveraging this new KD method, the plenty of knowledge encoded in a large "teacher" BERT can be effectively transferred to a small "student" Tiny-BERT. Then, we introduce a new two-stage learning framework for TinyBERT, which performs Transformer distillation at both the pretraining and task-specific learning stages. This framework ensures that TinyBERT can capture the general-domain as well as the task-specific knowledge in BERT. TinyBERT 41 with 4 layers is empirically effective and achieves more than 96.8% the performance of its teacher BERT BASE on GLUE benchmark, while being 7.5x smaller and 9.4x faster on inference. TinyBERT 4 is also significantly better than 4-layer state-of-the-art baselines on BERT distillation, with only ∼28% parameters and ∼31% inference time of them. Moreover, TinyBERT 6 with 6 layers performs on-par with its teacher BERT BASE .

We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models (Peters et al., 2018a;Radford et al., 2018), BERT is designed to pretrain deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers. As a result, the pre-trained BERT model can be finetuned with just one additional output layer to create state-of-the-art models for a wide range of tasks, such as question answering and language inference, without substantial taskspecific architecture modifications.BERT is conceptually simple and empirically powerful. It obtains new state-of-the-art results on eleven natural language processing tasks, including pushing the GLUE score to 80.5% (7.7% point absolute improvement), MultiNLI accuracy to 86.7% (4.6% absolute improvement), SQuAD v1.1 question answering Test F1 to 93.2 (1.5 point absolute improvement) and SQuAD v2.0 Test F1 to 83.1 (5.1 point absolute improvement).

Recent progress in pre-trained neural language models has significantly improved the performance of many natural language processing (NLP) tasks. In this paper we propose a new model architecture DeBERTa (Decoding-enhanced BERT with disentangled attention) that improves the BERT and RoBERTa models using two novel techniques. The first is the disentangled attention mechanism, where each word is represented using two vectors that encode its content and position, respectively, and the attention weights among words are computed using disentangled matrices on their contents and relative positions, respectively. Second, an enhanced mask decoder is used to incorporate absolute positions in the decoding layer to predict the masked tokens in model pre-training. In addition, a new virtual adversarial training method is used for fine-tuning to improve models' generalization. We show that these techniques significantly improve the efficiency of model pre-training and the performance of both natural language understand (NLU) and natural langauge generation (NLG) downstream tasks. Compared to RoBERTa-Large, a DeBERTa model trained on half of the training data performs consistently better on a wide range of NLP tasks, achieving improvements on MNLI by +0.9% (90.2% vs. 91.1%), on SQuAD v2.0 by +2.3% (88.4% vs. 90.7%) and RACE by +3.6% (83.2% vs. 86.8%). Notably, we scale up DeBERTa by training a larger version that consists of 48 Transform layers with 1.5 billion parameters. The significant performance boost makes the single DeBERTa model surpass the human performance on the SuperGLUE benchmark (Wang et al., 2019a) for the first time in terms of macro-average score (89.9 versus 89.8), and the ensemble DeBERTa model sits atop the SuperGLUE leaderboard as of January 6, 2021, outperforming the human baseline by a decent margin (90.3 versus 89.8). The pre-trained DeBERTa models and the source code were released at: https://github.com/microsoft/DeBERTa 1 .

We present Pre-trained Machine Reader (PMR), a novel method to retrofit Pre-trained Language Models (PLMs) into Machine Reading Comprehension (MRC) models without acquiring labeled data. PMR is capable of resolving the discrepancy between model pre-training and downstream fine-tuning of existing PLMs, and provides a unified solver for tackling various extraction tasks. To achieve this, we construct a large volume of general-purpose and high-quality MRC-style training data with the help of Wikipedia hyperlinks and design a Wiki Anchor Extraction task to guide the MRC-style pre-training process. Although conceptually simple, PMR is particularly effective in solving extraction tasks including Extractive Question Answering and Named Entity Recognition, where it shows tremendous improvements over previous approaches especially under low-resource settings. Moreover, viewing sequence classification task as a special case of extraction task in our MRC formulation, PMR is even capable to extract high-quality rationales to explain the classification process, providing more explainability of the predictions.

In the last year, new models and methods for pretraining and transfer learning have driven striking performance improvements across a range of language understanding tasks. The GLUE benchmark, introduced a little over one year ago, offers a single-number metric that summarizes progress on a diverse set of such tasks, but performance on the benchmark has recently surpassed the level of non-expert humans, suggesting limited headroom for further research. In this paper we present SuperGLUE, a new benchmark styled after GLUE with a new set of more difficult language understanding tasks, a software toolkit, and a public leaderboard. SuperGLUE is available at super.gluebenchmark.com.

立场检测旨在确定文本的作者是否赞成，反对或中立。这项任务的主要挑战是两个方面的：由于不同目标以及缺乏目标的上下文信息而产生的几乎没有学习。现有作品主要通过设计基于注意力的模型或引入嘈杂的外部知识来解决第二期，而第一个问题仍未探索。在本文中，受到预训练的语言模型（PLM）的潜在能力（PLM）的启发，我们建议介绍基于立场检测的及时基于迅速的微调。 PLM可以为目标提供基本的上下文信息，并通过提示启用几次学习。考虑到目标在立场检测任务中的关键作用，我们设计了目标感知的提示并提出了一种新颖的语言。我们的语言器不会将每个标签映射到具体单词，而是将每个标签映射到矢量，并选择最能捕获姿势与目标之间相关性的标签。此外，为了减轻通过单人工提示来处理不同目标的可能缺陷，我们建议将信息从多个提示中学到的信息提炼。实验结果表明，我们提出的模型在全数据和少数场景中的表现出色。

多语言语言模型（\ mllms），如mbert，xlm，xlm-r，\ textit {etc。}已成为一种可行的选择，使预先估计到大量语言的力量。鉴于他们的成功在零射击转移学习中，在（i）建立更大的\ mllms〜覆盖了大量语言（ii）创建覆盖更广泛的任务和语言来评估的详尽工作基准mllms〜（iii）分析单音零点，零拍摄交叉和双语任务（iv）对Monolingual的性能，了解\ mllms〜（v）增强（通常）学习的通用语言模式（如果有的话）有限的容量\ mllms〜以提高他们在已见甚至看不见语言的表现。在这项调查中，我们审查了现有的文学，涵盖了上述与\ MLLMS有关的广泛研究领域。根据我们的调查，我们建议您有一些未来的研究方向。

预先接受的语言模型实现了最先进的导致各种自然语言处理（NLP）任务。 GPT-3表明，缩放预先训练的语言模型可以进一步利用它们的巨大潜力。最近提出了一个名为Ernie 3.0的统一框架，以预先培训大型知识增强型号，并培训了具有10亿参数的模型。 Ernie 3.0在各种NLP任务上表现出最先进的模型。为了探讨缩放的表现，我们培养了百卢比的3.0泰坦参数型号，在PaddlePaddle平台上有高达260亿参数的泰坦。此外，我们设计了一种自我监督的对抗性损失和可控语言建模损失，以使ERNIE 3.0 TITAN产生可信和可控的文本。为了减少计算开销和碳排放，我们向Ernie 3.0泰坦提出了一个在线蒸馏框架，教师模型将同时教授学生和培训。埃塞尼3.0泰坦是迄今为止最大的中国密集预训练模型。经验结果表明，Ernie 3.0泰坦在68个NLP数据集中优于最先进的模型。

Sequence-to-sequence (seq2seq) learning is a popular fashion for large-scale pretraining language models. However, the prior seq2seq pretraining models generally focus on reconstructive objectives on the decoder side and neglect the effect of encoder-side supervision, which we argue may lead to sub-optimal performance. To verify our hypothesis, we first empirically study the functionalities of the encoder and decoder in seq2seq pretrained language models, and find that the encoder takes an important but under-exploitation role than the decoder regarding the downstream performance and neuron activation. Therefore, we propose an encoding-enhanced seq2seq pretraining strategy, namely E2S2, which improves the seq2seq models via integrating more efficient self-supervised information into the encoders. Specifically, E2S2 adopts two self-supervised objectives on the encoder side from two aspects: 1) locally denoising the corrupted sentence (denoising objective); and 2) globally learning better sentence representations (contrastive objective). With the help of both objectives, the encoder can effectively distinguish the noise tokens and capture high-level (i.e. syntactic and semantic) knowledge, thus strengthening the ability of seq2seq model to accurately achieve the conditional generation. On a large diversity of downstream natural language understanding and generation tasks, E2S2 dominantly improves the performance of its powerful backbone models, e.g. BART and T5. For example, upon BART backbone, we achieve +1.1% averaged gain on the general language understanding evaluation (GLUE) benchmark and +1.75% F_0.5 score improvement on CoNLL2014 dataset. We also provide in-depth analyses to show the improvement stems from better linguistic representation. We hope that our work will foster future self-supervision research on seq2seq language model pretraining.

Recent advances in NLP are brought by a range of large-scale pretrained language models (PLMs). These PLMs have brought significant performance gains for a range of NLP tasks, circumventing the need to customize complex designs for specific tasks. However, most current work focus on finetuning PLMs on a domain-specific datasets, ignoring the fact that the domain gap can lead to overfitting and even performance drop. Therefore, it is practically important to find an appropriate method to effectively adapt PLMs to a target domain of interest. Recently, a range of methods have been proposed to achieve this purpose. Early surveys on domain adaptation are not suitable for PLMs due to the sophisticated behavior exhibited by PLMs from traditional models trained from scratch and that domain adaptation of PLMs need to be redesigned to take effect. This paper aims to provide a survey on these newly proposed methods and shed light in how to apply traditional machine learning methods to newly evolved and future technologies. By examining the issues of deploying PLMs for downstream tasks, we propose a taxonomy of domain adaptation approaches from a machine learning system view, covering methods for input augmentation, model optimization and personalization. We discuss and compare those methods and suggest promising future research directions.

Transformer-based models have pushed state of the art in many areas of NLP, but our understanding of what is behind their success is still limited. This paper is the first survey of over 150 studies of the popular BERT model. We review the current state of knowledge about how BERT works, what kind of information it learns and how it is represented, common modifications to its training objectives and architecture, the overparameterization issue and approaches to compression. We then outline directions for future research.

我们提出了Patron，这是一种新方法，它使用基于及时的不确定性估计，用于在冷启动场景下进行预训练的语言模型进行微调的数据选择，即，没有初始标记的数据可用。在顾客中，我们设计（1）一种基于迅速的不确定性传播方法来估计数据点的重要性和（2）分区 - 然后 - 剥离（PTR）策略，以促进对注释的样品多样性。六个文本分类数据集的实验表明，赞助人的表现优于最强的冷启动数据选择基准，高达6.9％。此外，仅具有128个标签，顾客分别基于香草微调和及时的学习，获得了91.0％和92.1％的全面监督性能。我们的赞助人实施可在\ url {https://github.com/yueyu1030/patron}上获得。

In this work, we explore "prompt tuning," a simple yet effective mechanism for learning "soft prompts" to condition frozen language models to perform specific downstream tasks. Unlike the discrete text prompts used by GPT-3, soft prompts are learned through backpropagation and can be tuned to incorporate signals from any number of labeled examples. Our end-to-end learned approach outperforms GPT-3's few-shot learning by a large margin. More remarkably, through ablations on model size using T5, we show that prompt tuning becomes more competitive with scale: as models exceed billions of parameters, our method "closes the gap" and matches the strong performance of model tuning (where all model weights are tuned). This finding is especially relevant because large models are costly to share and serve and the ability to reuse one frozen model for multiple downstream tasks can ease this burden. Our method can be seen as a simplification of the recently proposed "prefix tuning" of Li and Liang (2021) and we provide a comparison to this and other similar approaches. Finally, we show that conditioning a frozen model with soft prompts confers benefits in robustness to domain transfer and enables efficient "prompt ensembling." * Work done as a Google AI Resident.

Transfer learning, where a model is first pre-trained on a data-rich task before being finetuned 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.

快速学习已成为现代自然语言处理的新范式，它直接适应培训的语言模型（PLMS）到$ CLOZE $ -Style预测，自回归建模或序列到序列生成，从而导致各种任务的表现。但是，尚未提出及时学习的标准实施框架，以及大多数现有的及时学习码条，通常是不受管制的，仅为特定方案提供有限的实现。由于有许多细节，例如模板策略，初始化策略和语言化策略等，因此需要在快速学习中考虑，从业者面临障碍，以便快速调整所需的迅速学习方法到他们的应用程序。在本文中，我们展示了{OpenPrompt}，一个统一的易于使用的工具包，可以通过PLMS快速学习。 OpenPrompt是一项研究型框架，配备了效率，模块化和可扩展性，其组合性允许自由地将不同的PLMS，任务格式和提示模块组合在统一的范例中。用户可以宽松地部署快速学习框架，并在没有约束的情况下在不同的NLP任务上评估它们的泛化。 OpenPrompt在{\ url {https://github.com/thunlp/openprompt}}上公开发布。

提示调整（PT）是一个有前途的参数高效的方法，可以利用极大的预先培训的语言模型（PLM），它可以通过仅调整几个软提示来实现与全参数微调的可比性。但是，与微调相比，PT经验需要更多的培训步骤。为了探索我们通过重用培训的软提示和分享知识来提高PT的效率，我们经验探讨了在不同任务和模型中的软提示的可转换性。在交叉任务传输中，我们发现训练有素的软提示可以转移到类似的任务并初始化PT，以加速培训并提高性能。此外，为了探讨影响的因素，提示跨任务的可转移性，我们调查如何测量提示相似性，并发现激活神经元的重叠率与可转移性高度相关。在跨模型传输中，我们探索如何将PLM的提示投影到另一个PLM并成功培训了一种可以在类似任务上实现非琐碎的传输性能的投影仪。但是，使用预计提示初始化PT不起作用，这可能是由优化偏好和PLMS高冗余引起的。我们的研究结果表明，具有知识转移的改善PT是可能的并且有希望的，而提示的交叉任务转移性通常比跨模型转移性更好。

预训练的语言模型（PLM）在自然语言生成（NLG）任务中取得了显着的成功。到目前为止，大多数PLM都使用大型一般语料库以无监督的方式进行了预培训。同时，与无监督的模型相比，预先训练的模型越来越多地显示出较低的数据表现出色。受监督预训练的成功的激励，我们提出了自然语言生成的多任务监督预训练（MVP）。为了预先培训文本生成模型MVP，我们从七个生成任务中收集了45个数据集的标记预训练语料库。对于每个任务，我们进一步预先训练特定的软提示，以刺激执行特定任务的模型能力。广泛的实验证明了我们在许多NLG任务中有监督的预训练的有效性，并且我们的一般方法在17个数据集中的12个中实现了最先进的性能。

提示方法被认为是几次自然语言处理的关键进展之一。最近对基于离散令牌的``硬提示''转移到连续``软提示''的最新研究，这些提示将可学习的向量用作伪提示代币并实现更好的性能。尽管显示出有希望的前景，但观察到这些软宣传的方法在很大程度上依赖良好的初始化来生效。不幸的是，获得软提示的完美初始化需要了解内在语言模型的工作和精心设计，这绝非易事，必须从头开始重新启动每个新任务。为了解决此问题，我们提出了一种称为Metaprompting的广义软提示方法，该方法采用了良好认可的模型 - 静态元学习算法，以自动找到更好的及时初始化，从而快速适应新的促进任务。问题并在四个不同的数据集上带来了显着改善（1次设置的准确性提高了6分），从而实现了新的最新性能。