United Arab Emirates

Building a question answering (QA) model with less annotation costs can be achieved by utilizing active learning (AL) training strategy. It selects the most informative unlabeled training data to update the model effectively. Acquisition functions for AL are used to determine how informative each training example is, such as uncertainty or diversity based sampling. In this work, we propose a perturbation-based active learning acquisition strategy and demonstrate it is more effective than existing commonly used strategies for the QA task.

EMNLP 2022

Perturbation-based Active Learning for Question Answering

reading comprehension

question answering

active learning

workshop paper

## Welcome to EMNLP 2022!
I am delighted to welcome you to EMNLP 2022! I believe this conference has been complicated beyond any precedent. Over the past year, it’s been thrilling to see the organization team approach each new puzzle with creativity and enthusiasm. We hope that those participating in Abu Dhabi as well as those joining remotely will leave the conference feeling newly inspired by the program and newly connected to our ever-growing community. Following EMNLP 2021 and major NLP conferences since, EMNLP 2022 is “hybrid,” serving both virtual and in-person participants.

Our key innovations for EMNLP 2022 include:

* EMNLP 2022 is “hybrid” in a second sense, as well: we allowed both direct and rolling review paper submissions, building on the pilot experiment of EMNLP 2021, which considered a small number of ARR submissions. 
* Familiar from NAACL but new to EMNLP, we’ve added an industry track.
* During the conference, “portals” will link virtual poster sessions to in-person conference participants during poster sessions each day.
* The first ACL-family conference in the United Arab Emirates.

 *Message from Noah A. Smith, University of Washington and Allen Institute for AI, Seattle, Washington, USA* 
***EMNLP 2022 General Chair***
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/9eec7d4a287ee18c278b08229290aa83.png)](https://drive.google.com/file/d/1OlPv6QBeo62VVTughj2jkiLeyHd1WnUt/view)
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/a3db7a768409f05192210d98601edb25.png)](https://emnlp2022.rocket.chat/)

To access this site you need to register. Please register [here](https://2022.emnlp.org/registration/).

Register here

Welcome!
EMNLP 2022 will take place in Abu Dhabi from December 7th to December 11th, 2022. And it will be held in hybrid mode, both online and offline.

Identification of hate speech (HS) on social media platforms has become a Sisyphean task with the increment of web data availability and the human knack for discovering loopholes in the existing HS detection algorithms.This study conducts a comparative analysis of the merits of using two transformer approaches for HS detection: BART and RoBERTa. By considering popular benchmark data sets, we show that the BART outperforms RoBERTa for the monolingual classification of both finding hate and non-hate as well as classifying hate speech into categories.

Short Comparative Analysis on Pretrained BART and RoBERTa in
Detecting Hate Speech on YouTube and Reddit Platforms

These days, many people use social media as a source of information and medium of communication due to its easy to access, fast to disseminate and low cost platform. However, it also enables the wide propagation of fake news which causes economic, political, and social crises to the society. As a result, many researchers have been working towards detecting fake news. Most of the researches concerned on linguistic analysis of news content to identify its credibility, however fake news is also written intentionally to mislead users by mimicking true news. Beside this, Amharic is one of the under-resourced language that suffer from the benefits of fake news detection. To overcome the problem of fake news using content feature and under-resourced language, this study uses a feature fusion of linguistic and social context feature of the publisher information to detect Amharic fake news. The fusion-based experiment shows at least 94.13% and at most 98.7% relative error reduction over the content-based approaches.

Amharic Fake News Detection on Social Media Using Feature Fusion

The main goal of this work is to develop bi-directional machine translation between Amharic and Kistanigna. Due to an increase in the number of language users, to address the issues of the endangered of the Kistanigna language and to increases the content of the language in web, it is essential to develop machine translation between Kistanigna and Amharic language. In this study the translation is implemented using neural machine translation. We conduct the experiments using LSTM, Bi-LSTM, LSTM + attention, CNN + attention and Transformer encoder-decoder model. We considered training time, memory usage, and BLEU score when proposing an optimal model. Finally, we suggested the morpheme-based bidirectional machine translation using Transformer with BLEU scores of 21.31 and 22.40 for Amharic-Kistanigna and Kistanigna-Amharic translation respectively. The major weakness of the study is unavailability of enough dataset to conduct an extensive experiment. As a result, there is a need to prepare parallel corpora for conducting similar research.


Amharic-Kistangigna Bi-directional Machine Translation using Deep Learning

Lexical ambiguity is a pervasive feature of natural language, and a major difficulty in understanding language is selecting the intended meaning when more than one are possible. Despite this difficulty, many studies have found a processing advantage for words with multiple related meanings. There are mixed results for words with multiple unrelated meanings. Our work investigates whether contextualized word embeddings are able to capture human-like distinctions between senses and meanings and whether they can predict human behavior. We reanalyze data from previous experiments and find that our method does replicate the polyseme advantage and homonym disadvantage previously reported. Additionally, our contin- uous predictors are superior to binary distinctions derived from lexicographic resources.

Contextual Embeddings Can Distinguish Homonymy from Polysemy in a Human-Like Way

Garden path sentences (i.e. “the horse raced past the barn fell”) are sentences that readers initially incorrectly parse, requiring partial or total re-analysis of the sentence structure. Given human difficulty in parsing garden paths, we aim to compare transformer language models’ performance on these sentences. We assess a BERT question-answering model’s performance on comprehension questions based on garden path and control sentences. We then further investigate the semantic roles assigned to arguments of verbs in garden path and control sentences by utilizing a probe task to directly assess which semantic role(s) BERT assigns. We find that BERT has relatively low performance in question answering based on garden path contexts, and the model incorrectly assigns semantic roles, aligning for the most part with human performance.

The BERT Walked Down the Garden Path Assigned Semantic Roles

Automatic Speech Recognition (ASR) systems have become a very natural human-machine interaction in which it allows users to speak entries rather than punching numbers on a keypad. This study presents an ASR system for Amharic language that is one of the Ethiopian Languages. The current attempts on speech recognition systems require thousands of hours of transcribed speech dataset to reach adequate accuracy. However, the large majority of under-resourced spoken languages in general, Ethiopic languages in particular, have a very limited amount of labeled speech dataset. Thus, in this study, a self-supervised Transformer based Wave2Vec 2.0 approach has been conducted to build an ASR system. In order to pretrain the proposed model in an unsupervised approach, a total of more than 200 hours unlabeled speech have been collected. In addition, a total of 30 minutes of labeled speech dataset has been prepared. Then, a Wave2Vec model that is initialized with weighted parameters has been pre-trained on the unlabeled speech 
dataset. Then the pretrained model has been fine-tuned using a small amount of labeled speech dataset. A standard Word Error Rate (WER) evaluation metric has been used to evaluate the fine-tuned multilingual ASR model and it has 
shown a clear significant result.

Automatic Speech Recognition for Amharic Language using Self-Supervised

The main goal of this work is to develop sentence level automatic speech segmentation system for Amharic. Sentence segmentation is a process of identifying the end of a sentence. In this study, sentence segmentation system is 
implemented in to two approaches. In the first approach, we used an automatic tool for segmenting and labeling of Amharic speech data. Acoustic model is created using 
speech and their text scripts and compiling them into a statistical representation of sounds which makeup words. This is done through HMM modeling. The approach one automatic speech segmentation system is done by forced alignment. In this approach we used rule-based and AdaBoost to discriminate the true boundaries from false. In the second approach, we extracted prosodic features directly 
from speech waveform and also statistical method, AdaBoost, is used.

Prosody Based Automatic Speech segmentation for Amharic

The way sexual abuse is represented in the media can drive public opinion. We use NLP techniques to analyze representation of sexual abuse in New York Times (NYT) articles over three time periods, ending with the era of MeToo. We use computational methods to surface sentences in articles involving victims and perpetrators, which we annotate for victim and perpetrator demographic information. We use power and agency connotation frames to measure if and how ascribed power and agency of victims and perpetrators in news articles have changed over the years. We find that women are ascribed more agency in sentences describing abuse of women victims over time, and power ascribed to male perpetrators has decreased over time.

Before and beyond MeToo: Measuring changes in power and agency within sexual abuse news stories over time

In this work, we propose a flexible multimodal transformer architecture for classifying depression from social media posts containing text and images. We show that our model is able to surpass other methods by a large margin on a popular multimodal depression dataset from Twitter, obtaining state-of-the-art results of 95\% F1 score. Our model is flexible and can easily incorporate temporal information by manipulating the transformer positional encodings. Moreover, it can operate both on randomly sampled and unordered sets of posts, to be more robust to dataset noise, and on ordered collections of posts, accommodating the relative posting intervals without any major architectural modifications.

Detecting Depression on Twitter with a Time-Aware Multimodal Transformer

Out-of-Vocabulary (OOV) is a problem for Neural Machine Translation (NMT). OOV refers to words with a low occurrence in the training data or those absent from the training data. Word or phrase-based Data Augmentation (DA) techniques have been used to alleviate this. However, existing DA techniques have addressed only one of these OOV types and are limited to considering either syntactic or semantic constraints. We present a word/phrase replacement-based DA technique that considers both types of OOV, considering both syntactic and semantic properties of the words to guarantee fluency in the synthetic sentences. This technique was experimented with low resource language pair Sinhala-English.

Perturbation-based Active Learning for Question Answering

Next from EMNLP 2022

Short Comparative Analysis on Pretrained BART and RoBERTa in Detecting Hate Speech on YouTube and Reddit Platforms

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES