This paper mainly describes the dma submission to the TempoWiC task, which achieves a macro-F1 score of 77.05% and attains the first place in this task. We first explore the impact of different pre-trained language models. Then we adopt data cleaning, data augmentation, and adversarial training strategies to enhance the model generalization and robustness. For further improvement, we integrate POS information and word semantic representation using a Mixture-of-Experts (MoE) approach. The experimental results show that MoE can overcome the feature overuse issue and combine the context, POS, and word semantic features well. Additionally, we use a model ensemble method for the final prediction, which has been proven effective by many research works.
translated by 谷歌翻译
语言随着时间的流逝而演变,单词含义会发生相应的变化。在社交媒体中尤其如此,因为它的动态性质会导致语义转移的速度更快,这使得NLP模型在处理新内容和趋势方面具有挑战性。但是,专门解决这些社交平台动态性质的数据集和模型的数量很少。为了弥合这一差距,我们提出了Tempowic,这是一种新的基准,尤其是旨在加快基于社交媒体的含义转变的研究。我们的结果表明,即使对于最近发行的专门从事社交媒体的语言模型,Tempowic是一个具有挑战性的基准。
translated by 谷歌翻译
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).
translated by 谷歌翻译
我们介绍了Twhin-Bert,这是一种多语言语言模型,该模型在流行的社交网络Twitter上训练了内域数据。Twhin-bert与先前的预训练的语言模型有所不同,因为它不仅接受了基于文本的自学训练,而且还具有基于Twitter异质信息网络(TWHIN)中丰富社交活动的社会目标。我们的模型接受了70亿条推文的培训,涵盖了100多种不同的语言,为简短,嘈杂,用户生成的文本提供了有价值的表示形式。我们对各种多语言社会建议和语义理解任务进行评估,并证明了对既定的预训练的语言模型的大幅改进。我们将自由开放源代码Twhin-Bert和我们为研究社区提供的精心策划标签预测和社会参与基准数据集。
translated by 谷歌翻译
在本文中,我们介绍了TweetNLP,这是社交媒体中自然语言处理(NLP)的集成平台。TweetNLP支持一套多样化的NLP任务,包括诸如情感分析和命名实体识别的通用重点领域,以及社交媒体特定的任务,例如表情符号预测和进攻性语言识别。特定于任务的系统由专门用于社交媒体文本的合理大小的基于变压器的语言模型(尤其是Twitter)提供动力,无需专用硬件或云服务即可运行。TweetNLP的主要贡献是:(1)使用适合社会领域的各种特定于任务的模型,用于支持社交媒体分析的现代工具包的集成python库;(2)使用我们的模型进行无编码实验的交互式在线演示;(3)涵盖各种典型社交媒体应用的教程。
translated by 谷歌翻译
我们提出了视觉和启发性语言变压器(Vault)。 Vault是流行的视觉和语言变压器(VILT)的扩展,并提高了视觉和语言任务的性能,这些任务涉及比图像字幕更复杂的文本输入,同时对训练和推理效率的影响最小。重要的是,Vilt通过使用浅图像编码器实现了有效的培训和视觉和语言任务的推断。但是,它是在字幕和类似的数据集上鉴定的,在该数据集中,语言输入简单,文字和描述性,因此缺乏语言多样性。因此,当使用野外多媒体数据(例如多模式社交媒体数据(在我们的工作,Twitter)中)时,从字幕语言数据以及任务多样性都有显着转变,我们确实找到了证据表明该语言vilt的能力是缺乏的。保险库的关键见解是将大型语言模型(例如Bert)的输出表示传播到Vilt的语言输入。我们表明,这种策略在涉及更丰富的语言输入和情感构造的视觉和语言任务上大大改善了毒品,例如Twitter-2015,Twitter-2015,Twitter-2017,MVSA-Single和MVSA-Multiple,但落后于纯粹的推理任务之后作为彭博Twitter文本图像关系数据集。我们已经在https://github.com/gchochla/vault上发布了所有实验的代码。
translated by 谷歌翻译
转移学习已通过深度审慎的语言模型广泛用于自然语言处理,例如来自变形金刚和通用句子编码器的双向编码器表示。尽管取得了巨大的成功,但语言模型应用于小型数据集时会过多地适合,并且很容易忘记与分类器进行微调时。为了解决这个忘记将深入的语言模型从一个域转移到另一个领域的问题,现有的努力探索了微调方法,以减少忘记。我们建议DeepeMotex是一种有效的顺序转移学习方法,以检测文本中的情绪。为了避免忘记问题,通过从Twitter收集的大量情绪标记的数据来仪器进行微调步骤。我们使用策划的Twitter数据集和基准数据集进行了一项实验研究。 DeepeMotex模型在测试数据集上实现多级情绪分类的精度超过91%。我们评估了微调DeepeMotex模型在分类Emoint和刺激基准数据集中的情绪时的性能。这些模型在基准数据集中的73%的实例中正确分类了情绪。所提出的DeepeMotex-Bert模型优于BI-LSTM在基准数据集上的BI-LSTM增长23%。我们还研究了微调数据集的大小对模型准确性的影响。我们的评估结果表明,通过大量情绪标记的数据进行微调提高了最终目标任务模型的鲁棒性和有效性。
translated by 谷歌翻译
BERT,ROBERTA或GPT-3等复杂的基于注意力的语言模型的外观已允许在许多场景中解决高度复杂的任务。但是,当应用于特定域时,这些模型会遇到相当大的困难。诸如Twitter之类的社交网络就是这种情况,Twitter是一种不断变化的信息流,以非正式和复杂的语言编写的信息流,鉴于人类的重要作用,每个信息都需要仔细评估,即使人类也需要理解。通过自然语言处理解决该领域的任务涉及严重的挑战。当将强大的最先进的多语言模型应用于这种情况下,特定语言的细微差别用来迷失翻译。为了面对这些挑战,我们提出了\ textbf {bertuit},这是迄今为止针对西班牙语提出的较大变压器,使用Roberta Optimization进行了230m西班牙推文的大规模数据集进行了预培训。我们的动机是提供一个强大的资源,以更好地了解西班牙Twitter,并用于专注于该社交网络的应用程序,特别强调致力于解决该平台中错误信息传播的解决方案。对Bertuit进行了多个任务评估,并与M-Bert,XLM-Roberta和XLM-T进行了比较,该任务非常具有竞争性的多语言变压器。在这种情况下,使用应用程序显示了我们方法的实用性:一种可视化骗局和分析作者群体传播虚假信息的零击方法。错误的信息在英语以外的其他语言等平台上疯狂地传播,这意味着在英语说话之外转移时,变形金刚的性能可能会受到影响。
translated by 谷歌翻译
多字表达式(MWES)呈现单词组,其中整体的含义不是源于其部分的含义。处理MWE的任务在许多自然语言处理(NLP)应用中至关重要,包括机器翻译和术语提取。因此,检测MWE是一个流行的研究主题。在本文中,我们在检测MWES的任务中探索了最新的神经变压器。我们在数据集中凭经验评估了Semeval-2016任务10:检测最小的语义单元及其含义(DIMSUM)。我们表明,变压器模型的表现优于先前基于长期记忆(LSTM)的神经模型。该代码和预培训模型将免费提供给社区。
translated by 谷歌翻译
我们提出了一个多语言对抗训练模型,用于确定句子是否包含惯用表达式。鉴于该任务的关键挑战是注释数据的大小有限,我们的模型依赖于来自不同多语言最新变压器的语言模型(即多语言Bert和XLM-)的预训练的上下文表示。罗伯塔(Roberta),以及对抗训练,是一种进一步增强模型概括和鲁棒性的训练方法。我们的模型不依赖于任何人力制作的功能,知识库或其他数据集以外的其他数据集,我们的模型获得了竞争成果,并在子任务A(零射击)设置中排名第六,在子任务中排名第15位(单发)环境。
translated by 谷歌翻译
WSD (Word Sense Disambiguation) is the task of identifying which sense of a word is meant in a sentence or other segment of text. Researchers have worked on this task (e.g. Pustejovsky, 2002) for years but it's still a challenging one even for SOTA (state-of-the-art) LMs (language models). The new dataset, TempoWiC introduced by Loureiro et al. (2022b) focuses on the fact that words change over time. Their best baseline achieves 70.33% macro-F1. In this work, we use two different losses simultaneously to train RoBERTa-based classification models. We also improve our model by using another similar dataset to generalize better. Our best configuration beats their best baseline by 4.23% and reaches 74.56% macroF1.
translated by 谷歌翻译
Understanding customer feedback is becoming a necessity for companies to identify problems and improve their products and services. Text classification and sentiment analysis can play a major role in analyzing this data by using a variety of machine and deep learning approaches. In this work, different transformer-based models are utilized to explore how efficient these models are when working with a German customer feedback dataset. In addition, these pre-trained models are further analyzed to determine if adapting them to a specific domain using unlabeled data can yield better results than off-the-shelf pre-trained models. To evaluate the models, two downstream tasks from the GermEval 2017 are considered. The experimental results show that transformer-based models can reach significant improvements compared to a fastText baseline and outperform the published scores and previous models. For the subtask Relevance Classification, the best models achieve a micro-averaged $F1$-Score of 96.1 % on the first test set and 95.9 % on the second one, and a score of 85.1 % and 85.3 % for the subtask Polarity Classification.
translated by 谷歌翻译
在网络和社交媒体上生成的大量数据增加了检测在线仇恨言论的需求。检测仇恨言论将减少它们对他人的负面影响和影响。在自然语言处理(NLP)域中的许多努力旨在宣传仇恨言论或检测特定的仇恨言论,如宗教,种族,性别或性取向。讨厌的社区倾向于使用缩写,故意拼写错误和他们的沟通中的编码词来逃避检测,增加了讨厌语音检测任务的更多挑战。因此,词表示将在检测仇恨言论中发挥越来越关的作用。本文研究了利用基于双向LSTM的深度模型中嵌入的域特定词语的可行性,以自动检测/分类仇恨语音。此外,我们调查转移学习语言模型(BERT)对仇恨语音问题作为二进制分类任务。实验表明,与双向LSTM基于LSTM的深层模型嵌入的域特异性词嵌入了93%的F1分数,而BERT在可用仇恨语音数据集中的组合平衡数据集上达到了高达96%的F1分数。
translated by 谷歌翻译
由于BERT出现,变压器语言模型和转移学习已成为自然语言理解任务的最先进。最近,一些作品适用于特定领域的预训练,专制模型,例如科学论文,医疗文件等。在这项工作中,我们呈现RoberTuito,用于西班牙语中的用户生成内容的预先训练的语言模型。我们在西班牙语中培训了罗伯特托5亿推文。关于涉及用户生成文本的4个任务的基准测试显示,罗伯特托多于西班牙语的其他预先接受的语言模型。为了帮助进一步研究,我们将罗伯特多公开可在HuggingFace Model Hub上提供。
translated by 谷歌翻译
作为世界上第四大语言家庭,Dravidian语言已成为自然语言处理(NLP)的研究热点。虽然Dravidian语言包含大量语言,但有相对较少的公众可用资源。此外,文本分类任务是自然语言处理的基本任务,如何将其与Dravidian语言中的多种语言相结合,仍然是Dravidian自然语言处理的主要困难。因此,为了解决这些问题,我们为Dravidian语言提出了一个多语言文本分类框架。一方面,该框架使用Labse预先训练的模型作为基础模型。针对多任务学习中文本信息偏见的问题,我们建议使用MLM策略选择语言特定的单词,并使用对抗训练来扰乱它们。另一方面,鉴于模型无法识别和利用语言之间的相关性的问题,我们进一步提出了一种特定于语言的表示模块,以丰富模型的语义信息。实验结果表明,我们提出的框架在多语言文本分类任务中具有重要性能,每个策略实现某些改进。
translated by 谷歌翻译
讽刺可以被定义为说或写讽刺与一个人真正想表达的相反,通常是为了侮辱,刺激或娱乐某人。由于文本数据中讽刺性的性质晦涩难懂,因此检测到情感分析研究社区的困难和非常感兴趣。尽管讽刺检测的研究跨越了十多年,但最近已经取得了一些重大进步,包括在多模式环境中采用了无监督的预训练的预训练的变压器,并整合了环境以识别讽刺。在这项研究中,我们旨在简要概述英语计算讽刺研究的最新进步和趋势。我们描述了与讽刺有关的相关数据集,方法,趋势,问题,挑战和任务,这些数据集,趋势,问题,挑战和任务是无法检测到的。我们的研究提供了讽刺数据集,讽刺特征及其提取方法以及各种方法的性能分析,这些表可以帮助相关领域的研究人员了解当前的讽刺检测中最新实践。
translated by 谷歌翻译
Due to their crucial role in all NLP, several benchmarks have been proposed to evaluate pretrained language models. In spite of these efforts, no public benchmark of diverse nature currently exists for evaluation of Arabic. This makes it challenging to measure progress for both Arabic and multilingual language models. This challenge is compounded by the fact that any benchmark targeting Arabic needs to take into account the fact that Arabic is not a single language but rather a collection of languages and varieties. In this work, we introduce ORCA, a publicly available benchmark for Arabic language understanding evaluation. ORCA is carefully constructed to cover diverse Arabic varieties and a wide range of challenging Arabic understanding tasks exploiting 60 different datasets across seven NLU task clusters. To measure current progress in Arabic NLU, we use ORCA to offer a comprehensive comparison between 18 multilingual and Arabic language models. We also provide a public leaderboard with a unified single-number evaluation metric (ORCA score) to facilitate future research.
translated by 谷歌翻译
Automated offensive language detection is essential in combating the spread of hate speech, particularly in social media. This paper describes our work on Offensive Language Identification in low resource Indic language Marathi. The problem is formulated as a text classification task to identify a tweet as offensive or non-offensive. We evaluate different mono-lingual and multi-lingual BERT models on this classification task, focusing on BERT models pre-trained with social media datasets. We compare the performance of MuRIL, MahaTweetBERT, MahaTweetBERT-Hateful, and MahaBERT on the HASOC 2022 test set. We also explore external data augmentation from other existing Marathi hate speech corpus HASOC 2021 and L3Cube-MahaHate. The MahaTweetBERT, a BERT model, pre-trained on Marathi tweets when fine-tuned on the combined dataset (HASOC 2021 + HASOC 2022 + MahaHate), outperforms all models with an F1 score of 98.43 on the HASOC 2022 test set. With this, we also provide a new state-of-the-art result on HASOC 2022 / MOLD v2 test set.
translated by 谷歌翻译
本文介绍了我们对SMM4H 2022共享任务的提交,内容涉及自我报告的亲密伴侣暴力在Twitter上(英语)。这项任务的目的是准确确定给定推文的内容是否证明了某人报告自己的亲密伴侣暴力经历。提交的系统是五个罗伯塔模型组成的合奏,每个模型各自在验证数据集上由各自的F1分数加权。该系统的性能比基线要好13%,并且是该共享任务的总体性能最佳系统。
translated by 谷歌翻译
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 .
translated by 谷歌翻译