EMNLP2020 | 近期必读Natural Language Inference精选论文

**AMiner平台**由清华大学计算机系研发，拥有我国完全自主知识产权。平台包含了超过2.3亿学术论文/专利和1.36亿学者的科技图谱，提供学者评价、专家发现、智能指派、学术地图等科技情报专业化服务。系统2006年上线，吸引了全球220个国家/地区1000多万独立IP访问，数据下载量230万次，年度访问量超过1100万，成为学术搜索和社会网络挖掘研究的重要数据和实验平台。

AMiner平台：https://www.aminer.cn

导语：自然语言推理（Natural Language Inference，NLI）也被称为文本蕴含识别，是自然语言处理中最重要的问题之一，它需要推断两个给定句子之间的逻辑关系。该任务给出了两个句子，分别称为前提和假设，目标是确定它们之间的逻辑关系是必要的、中立的或矛盾的。目前，基于深度学习的文本匹配、预训练的语言模型等多种自然语言处理的方法，都在该领域中大放异彩。

根据AMiner-EMNLP2020词云图和论文可以看出，Natural Language Inference相关工作在本次会议中也有许多不凡的工作，下面我们一起看看Natural Language Inference主题的相关论文。

1.论文名称：Discriminative Nearest Neighbor Few-Shot Intent Detection by Transferring Natural Language Inference.

论文链接：https://www.aminer.cn/pub/5f7fe6d80205f07f689731b7?conf=emnlp2020

作者：Jianguo Zhang, Kazuma Hashimoto, Wenhao Liu, Chien-Sheng Wu, Yao Wan, Philip Yu, Richard Socher, Caiming Xiong

简介：

• Intent detection is one of the core components when building
  goal-oriented dialog systems.

• The authors have presented a simple yet efficient nearest-neighbor
  classification model to detect user intents and OOS intents.

• It includes paired encoding and discriminative training to model
  relations between the input and example utterances.

• Experimental results show superior performance of the method on a
  large-scale multi-domain intent detection dataset with OOS.

2.论文名称：Discriminatively-Tuned Generative Classifiers for Robust Natural Language Inference.

论文链接：https://www.aminer.cn/pub/5f7fe6d80205f07f689731b8?conf=emnlp2020

作者：Xiaoan Ding, Tianyu Liu, Baobao Chang, Zhifang Sui, Kevin Gimpel

简介：

• Natural language inference (NLI) is the task of identifying the relationship between two fragments of text, called the premise and the hypothesis.

• The authors proposed GenNLI, a discriminatively-finetuned generative classifier for NLI tasks, and empirically characterized its performance by comparing it to discriminative models and pretrained models.

• The authors conducted extensive experiments with GenNLI, showing its robustness across challenging empirical conditions.

• The authors showed its ability to generate hypotheses given premises and particular labels.

3.论文名称：Unsupervised Natural Language Inference via Decoupled Multimodal Contrastive Learning.

论文链接：https://www.aminer.cn/pub/5f7fe6d80205f07f68973341?conf=emnlp2020

作者：Wanyun Cui, Guangyu Zheng, Wei Wang

简介：

• The major flaw of previous multimodal SSL methods is that they use a joint encoder for representing the cross-modal correlations.

• This prevents them from integrating visual knowledge into the text encoder.

• The authors propose the multimodal aligned contrastive decoupled learning (MACD), which learns to represent visual knowledge while using only texts as inputs.

• In addition to the decoupled contrastive learning model, the authors propose two optimizations by adding the local structures into account, and by regularizing

4.论文名称：What Can We Learn from Collective Human Opinions on Natural Language Inference Data?

论文链接：https://www.aminer.cn/pub/5f7eef0891e011a5faf10028?conf=emnlp2020

作者：Yixin Nie, Xiang Zhou, Mohit Bansal

简介：

• Natural Language Understanding (NLU) evaluation plays a key role in benchmarking progress in natural language processing (NLP) research.

• While common practice in natural language evaluation compares the model prediction to the majority label, Section 5.4 questions the value of continuing such evaluation on current benchmarks as most of the unsolved examples are of low human agreement

• To address this concern, the authors suggest NLP models be evaluated against the collective human opinion distribution rather than one opinion aggregated from a set of opinions, especially on tasks which take a descriptivist approach10 to language and meaning, including NLI and common sense reasoning.

• This hints at the connections between human agreements and uncertainty estimation or calibration where machine learning models are required to produce the confidence value of their predictions, leading to important benefits in real-world applications

5.论文名称：ConjNLI: Natural Language Inference Over Conjunctive Sentences.

论文链接：https://www.aminer.cn/pub/5f7fe6d80205f07f6897318b?conf=emnlp2020

作者：Swarnadeep Saha, Yixin Nie, Mohit Bansal

简介：

• Recent years have seen significant progress in the task of Natural Language Inference (NLI) through the development of large-scale datasets like SNLI and MNLI.

• The authors presented CONJNLI, a new stress-test dataset for NLI in conjunctive sentences in the presence of negations and quantifiers and requiring diverse “boolean” and “nonboolean” inferences over conjuncts.

• The authors presented some initial solutions via adversarial training and a predicate-aware RoBERTa model, and achieved some reasonable performance gains on CONJNLI.

• The authors show limitations of the proposed methods, thereby encouraging future work on CONJNLI for better understanding of conjunctive semantics

更多EMNLP2020论文，可以关注公众号或者链接直达EMNLP2020专题，最前沿的研究方向和最全面的论文数据等你来~
EMNLP2020:https://www.aminer.cn/conf/emnlp2020