Effective Fine-Tuning Methods for Cross-lingual Adaptation

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 866–882 November 7–11, 2021. c⃝2021 Association for Computational Linguistics 866 Semantic Novelty Detection in Natural Language Descriptions Nianzu Ma†, Alexander Politowicz†, Sahisnu Mazumder†, Jiahua Chen†, Bing Liu†, Eric Robertson♮, Scott Grigsby♮ †Department of Computer Science, University of Illinois at Chicago, USA ♮PAR Government Systems Corporation, USA jingyima005@gmail.com, politow2@uic.edu {sahisnumazumder,jiahuaqy}@gmail.com liub@uic.edu, {eric_robertson,Scott_Grigsby}@partech.com Abstract This paper proposes to study a fine-grained se- mantic novelty detection task, which can be il- lustrated with the following example. It is nor- mal that a person walks a dog in the park, but if someone says “A man is walking a chicken in the park,” it is novel. Given a set of natu- ral language descriptions of normal scenes, we want to identify descriptions of novel scenes. We are not aware of any existing work that solves the problem. Although existing novelty or anomaly detection algorithms are applica- ble, since they are usually topic-based, they perform poorly on our fine-grained semantic novelty detection task. This paper proposes an effective model (called GAT-MA) to solve the problem and also contributes a new dataset. Experimental evaluation shows that GAT-MA outperforms 11 baselines by large margins. 1 Introduction Novelty or anomaly detection has been an impor- tant research topic since 1970s (Barnett and Lewis, 1994) due to numerous applications (Chalapathy et al., 2018; Pang et al., 2021). Recently, it has also become important for natural language processing (NLP). Many researchers have studied the problem in the text classification setting (Fei and Liu, 2016; Shu et al., 2017; Xu et al., 2019; Lin and Xu, 2019; Zheng et al., 2020). However, these text novelty classifiers are mainly coarse-grained, working at the document or topic level. Given a text document, their goal is to detect whether the text belongs to a known class or unknown class. This paper introduces a new text novelty detec- tion problem - fine-grained semantic novelty detec- tion. Specifically, given a text description d, we detect whether d represents a semantically novel fact or not. This work considers text data that de- scribe scenes of real-world phenomena in natu- ral language (NL). In our daily lives, we observe different real-world phenomena (events, activities, situations, etc.) and often describe these obser- vations (referred as “scenes" onwards) in NL to others or write about them. It is quite natural to observe scenes that we have not seen before (i.e., novel scenes). For example, it is a common scene that “A person walks a dog in the park", but if someone says “A man is walking a chicken in the park", it is quite unexpected and novel. Detecting such semantic novelty requires complex concep- tual and semantic reasoning over text and thus, is a challenging NLP problem. Note that conceptu- ally, the judgement of the novelty of a scene is subjective and might differ from person to person. However, there are some scenes for which a major- ity of people have agreement about their novelty. A good …

Content truncated to 500 words