- Li, Shen, et al. "Analogical reasoning on chinese morphological and semantic relations." arXiv preprint arXiv:1805.06504 (2018).
- Qiu, Yuanyuan, et al. "Revisiting correlations between intrinsic and extrinsic evaluations of word embeddings." Chinese Computational Linguistics and Natural Language Processing Based on Naturally Annotated Big Data. Springer, Cham, 2018. 209-221.
- Jeffrey Pennington, Richard Socher, and Christopher D. Manning. 2014. GloVe: Global Vectors for Word Representation.
- T. Mikolov, E. Grave, P. Bojanowski, C. Puhrsch, A. Joulin. Advances in Pre-Training Distributed Word Representations
- T. Mikolov, E. Grave, P. Bojanowski, C. Puhrsch, A. Joulin. Advances in Pre-Training Distributed Word Representations.
在自然语言处理(NLP)领域中,语言模型预训练方法在多项NLP任务上都获得了不错的提升,广泛受到了各界的关注。在这里主要是提供了目前两种语言模型,一种是RNNLM模型,通过RNN网络来进行序列任务的预测;另外一种是ELMo模型,以双向 LSTM 为网路基本组件,以 Language Model 为训练目标,通过预训练得到通用的语义表示。
| [ERNIE-GEN(An Enhanced Multi-Flow Pre-training and Fine-tuning Framework for Natural Language Generation)](../examples/text_generation/ernie-gen) | ERNIE-GEN是百度发布的生成式预训练模型,通过Global-Attention的方式解决训练和预测曝光偏差的问题,同时使用Multi-Flow Attention机制来分别进行Global和Context信息的交互,同时通过片段生成的方式来增加语义相关性。|
机器翻译是计算语言学的一个分支,是人工智能的终极目标之一,具有重要的科学研究价值。在机器翻译的任务上,提供了两大类模型,一类是传统的 Sequence to Sequence任务,简称Seq2Seq,通过RNN类模型进行编码,解码;另外一类是Transformer类模型,通过Self-Attention机制来提升Encoder和Decoder的效果,Transformer模型的具体信息可以参考论文, [Attention Is All You Need](https://arxiv.org/abs/1706.03762)。下面是具体的模型信息。
| [Dialogue General Understanding](../examples/dialogue/dgu) | 提供基于BERT通用对话理解模型,通过对文本分类、序列标注等操作就可以完成对话中的意图识别,行文识别,状态跟踪等理解任务。|
| [PLATO-2](../examples/dialogue/plato-2) | 百度自研领先的开放域对话预训练模型。[PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning](https://arxiv.org/abs/2006.16779) |
| [ERNIE-GEN(An Enhanced Multi-Flow Pre-training and Fine-tuning Framework for Natural Language Generation)](./text_generation/ernie-gen) | ERNIE-GEN是百度发布的生成式预训练模型,通过Global-Attention的方式解决训练和预测曝光偏差的问题,同时使用Multi-Flow Attention机制来分别进行Global和Context信息的交互,同时通过片段生成的方式来增加语义相关性。|
| [ERNIE-GEN(An Enhanced Multi-Flow Pre-training and Fine-tuning Framework for Natural Language Generation)](../examples/text_generation/ernie-gen) | ERNIE-GEN是百度发布的生成式预训练模型,通过Global-Attention的方式解决训练和预测曝光偏差的问题,同时使用Multi-Flow Attention机制来分别进行Global和Context信息的交互,同时通过片段生成的方式来增加语义相关性。|
| [Dialogue General Understanding](./dialogue/dgu) | 提供基于BERT通用对话理解模型,通过对文本分类、序列标注等操作就可以完成对话中的意图识别,行文识别,状态跟踪等理解任务。|
| [PLATO-2](./dialogue/plato-2) | 百度自研领先的开放域对话预训练模型。[PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning](https://arxiv.org/abs/2006.16779) |
| [Dialogue General Understanding](../examples/dialogue/dgu) | 提供基于BERT通用对话理解模型,通过对文本分类、序列标注等操作就可以完成对话中的意图识别,行文识别,状态跟踪等理解任务。|
| [PLATO-2](../examples/dialogue/plato-2) | 百度自研领先的开放域对话预训练模型。[PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning](https://arxiv.org/abs/2006.16779) |