United States

Domain divergences are a tool to measure domain shifts. Given a plethora of divergence measures, In this talk, we provide a taxonomy of divergences consisting of three classes - 1) Geometric 2) Information Theoretic 3) Higher order measures. We detail their use cases in domain adaptation for 1) Data Selection 2) Learning domain invariant representations 3) Making decisions in the wild. While divergence measures have a practical use-case of predicting performance drops, it is still not clear which divergence measure is suitable for a NLP task. Also, most of the works in the literature calculate divergence measures based on simple frequency based word distributions or word embeddings. Does using contextual word representations have any advantage in predicting the drop in performance? We perform a large-scale empirical analysis consisting of more than 130 domain adaptation scenarios, 3 NLP tasks and 12 divergence measures. We find that divergences calculated using traditional word distributions still serve as strong baselines.

NAACL 2021

Domain Divergences: A Survey and Empirical Analysis

2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics.

**Whova App** 
Stay in touch with your fellow conference attendees via the [Whova App](https://whova.com/portal/webapp/nacon_202106/)

**Conference Structure**
https://2021.naacl.org/blog/conference-structure/

**Walkthrough video of how to NAACL 2021** 

Please take a moment to view this video explaining how to navigate the platform, attend sessions network with other attendees. 


<figure class="video_container">
  <iframe src="https://screencast-o-matic.com/watch/crhwbGVh3vx?v=6&ff=1&title=0&controls=1" width=640  height=350  frameborder="0" allowfullscreen="true"> </iframe>
</figure>

<br>
<br>

technical paper

Transformer is an attention-based neural network, which consists of two sublayers, namely, Self-Attention Network (SAN) and Feed-Forward Network (FFN). Existing research explores to enhance the two sublayers separately to improve the capability of Transformer for text representation. In this paper, we present a novel understanding of SAN and FFN as Mask Attention Networks (MANs) and show that they are two special cases of MANs with static mask matrices. However, their static mask matrices limit the capability for localness modeling in text representation learning. We therefore introduce a new layer named dynamic mask attention network (DMAN) with a learnable mask matrix which is able to model localness adaptively. To incorporate advantages of DMAN, SAN, and FFN, we propose a sequential layered structure to combine the three types of layers. Extensive experiments on various tasks, including neural machine translation and text summarization demonstrate that our model outperforms the original Transformer.

Mask Attention Networks: Rethinking and Strengthen Transformer

We propose Lattice-BERT, a Chinese PLM enhanced with multi-granularity information from word lattices to directly, simultaneously, and interactively explore the representations of words and characters.

Lattice-BERT: Leveraging Multi-Granularity Representations in Chinese Pre-trained Language Models

Coarse-grained linguistic information, such as named entities or phrases, facilitates adequately representation learning in pre-training. Previous works mainly focus on extending the objective of BERT's Masked Language Modeling (MLM) from masking individual tokens to contiguous sequences of n tokens. We argue that such contiguously masking method neglects to model the intra-dependencies and inter-relation of coarse-grained linguistic information. As an alternative, we propose ERNIE-Gram, an explicitly n-gram masking method to enhance the integration of coarse-grained information into pre-training. In ERNIE-Gram, n-grams are masked and predicted directly using explicit n-gram identities rather than contiguous sequences of n tokens. Furthermore, ERNIE-Gram employs a generator model to sample plausible n-gram identities as optional n-gram masks and predict them in both coarse-grained and fine-grained manners to enable comprehensive n-gram prediction and relation modeling. We pre-train ERNIE-Gram on English and Chinese text corpora and fine-tune on 19 downstream tasks. Experimental results show that ERNIE-Gram outperforms previous pre-training models like XLNet and RoBERTa by a large margin, and achieves comparable results with state-of-the-art methods.

ERNIE-Gram: Pre-Training with Explicitly N-Gram Masked Language Modeling for Natural Language Understanding

"Word sense disambiguation (WSD) is a long-standing problem in natural language processing. One significant challenge in supervised all-words WSD is to classify among senses for a majority of words that lie in the long-tail distribution. For instance, 84% of the annotated words have less than 10 examples in the SemCor training data. This issue is more pronounced as the imbalance occurs in both word and sense distributions. In this work, we propose MetricWSD, a non-parametric few-shot learning approach to mitigate this data imbalance issue. By learning to compute distances among the senses of a given word through episodic training, MetricWSD transfers knowledge (a learned metric space) from high-frequency words to infrequent ones. MetricWSD constructs the training episodes tailored to word frequencies and explicitly addresses the problem of the skewed distribution, as opposed to mixing all the words trained with parametric models in previous work. Without resorting to any lexical resources, MetricWSD obtains strong performance against parametric alternatives, achieving a 75.1 F1 score on the unified WSD evaluation benchmark (Raganato et al., 2017b). Our analysis further validates that infrequent words and senses enjoy significant improvement.

Non-Parametric Few-Shot Learning for Word Sense Disambiguation

Word representations empowered with additional linguistic information have been widely studied and proved to outperform traditional embeddings. Current methods mainly focus on learning embeddings for words while embeddings of linguistic information (referred to as grain embeddings) are discarded after the learning. This work proposes a framework field embedding to jointly learn both word and grain embeddings by incorporating morphological, phonetic, and syntactical linguistic fields. The framework leverages an innovative fine-grained pipeline that integrates multiple linguistic fields and produces high-quality grain sequences for learning supreme word representations. A novel algorithm is also designed to learn embeddings for words and grains by capturing information that is contained within each field and that is shared across them. Experimental results of lexical tasks and downstream natural language processing tasks illustrate that our framework can learn better word embeddings and grain embeddings. Qualitative evaluations show grain embeddings effectively capture the semantic information.

Field Embedding: A Unified Grain-Based Framework for Word Representation

Automated metaphor detection is a challenging task to identify the metaphorical expression of words in a sentence. To tackle this problem, we adopt pre-trained contextualized models, e.g., BERT and RoBERTa. To this end, we propose a novel metaphor detection model, namely metaphor-aware late interaction over BERT. Our model not only leverages contextualized word representation but also benefits from linguistic metaphor identification theories to detect whether the target word is metaphorical. Our empirical results demonstrate that MelBERT outperforms several strong baselines on four benchmark datasets, i.e., VUA-18, VUA-20, MOH-X, and TroFi.

MelBERT: Metaphor Detection via Contextualized Late Interaction using Metaphorical Identification Theories

Fine-grained opinion mining (OM) has achieved increasing attraction in the natural language processing (NLP) community, which aims to find the opinion structures of ``Who expressed what opinions towards what'' in one sentence.

A Unified Span-Based Approach for Opinion Mining with Syntactic Constituents

The goal of stance detection is to identify whether the author of a text is in favor of, neutral or against a specific target. Despite substantial progress on this task, one of the remaining challenges is the scarcity of annotations. Data augmentation is commonly used to address annotation scarcity by generating more training samples. However, the augmented sentences that are generated by existing methods are either less diversified or inconsistent with the given target and stance label. In this work, we formulate the data augmentation of stance detection as a conditional masked language modeling task and augment the dataset by predicting the masked word conditioned on both its context and the auxiliary sentence that contains target and label information. Moreover, we propose another simple yet effective method that generates target-aware sentence by replacing a target mention with the other. Experimental results show that our proposed methods significantly outperforms previous augmentation methods on 11 targets.

Target-Aware Data Augmentation for Stance Detection

It's an about 9 minutes presentation of the paper: Target-specified Sequence Labeling with Multi-head Self-attention for Target-oriented Opinion Words Extraction, which introduces the task description, motivation, methodology and experiment results briefly. The topics are the target-oriented opinion words extraction and aspect-opinion pairs extraction.

Target-specified Sequence Labeling with Multi-head Self-attention for Target-oriented Opinion Words Extraction

Does syntax matter for Aspect-level Sentiment Classification？ No, the RoBERTa with an MLP layer is able to achieve SOTA or near SOTA performance in this task.

Domain Divergences: A Survey and Empirical Analysis

Downloads

Next from NAACL 2021

Mask Attention Networks: Rethinking and Strengthen Transformer

Similar lecture

MUSER: MUltimodal Stress detection using Emotion Recognition as an Auxiliary Task

Domain Divergences: A Survey and Empirical Analysis

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from NAACL 2021

Mask Attention Networks: Rethinking and Strengthen Transformer

Similar lecture

MUSER: MUltimodal Stress detection using Emotion Recognition as an Auxiliary Task

Downloads