Overview of AMALGUM – Large Silver Quality Annotations across English Genres

no code implementations • SCiL 2021 • Luke Gessler, Siyao Peng, Yang Liu, YIlun Zhu, Shabnam Behzad, Amir Zeldes

Sentence-level Feedback Generation for English Language Learners: Does Data Augmentation Help?

no code implementations • 18 Dec 2022 • Shabnam Behzad, Amir Zeldes, Nathan Schneider

In this paper, we present strong baselines for the task of Feedback Comment Generation for Writing Learning.

Data Augmentation

ELQA: A Corpus of Questions and Answers about the English Language

1 code implementation • 1 May 2022 • Shabnam Behzad, Keisuke Sakaguchi, Nathan Schneider, Amir Zeldes

We introduce a community-sourced dataset for English Language Question Answering (ELQA), which consists of more than 180k questions and answers on numerous topics about English language such as grammar, meaning, fluency, and etymology.

Answer Generation Question Answering

DisCoDisCo at the DISRPT2021 Shared Task: A System for Discourse Segmentation, Classification, and Connective Detection

1 code implementation • EMNLP (DISRPT) 2021 • Luke Gessler, Shabnam Behzad, Yang Janet Liu, Siyao Peng, YIlun Zhu, Amir Zeldes

This paper describes our submission to the DISRPT2021 Shared Task on Discourse Unit Segmentation, Connective Detection, and Relation Classification.

Classification Connective Detection +3

Team DoNotDistribute at SemEval-2020 Task 11: Features, Finetuning, and Data Augmentation in Neural Models for Propaganda Detection in News Articles

no code implementations • SEMEVAL 2020 • Michael Kranzlein, Shabnam Behzad, Nazli Goharian

This paper presents our systems for SemEval 2020 Shared Task 11: Detection of Propaganda Techniques in News Articles.

Data Augmentation Propaganda detection

AMALGUM -- A Free, Balanced, Multilayer English Web Corpus

1 code implementation • LREC 2020 • Luke Gessler, Siyao Peng, Yang Liu, YIlun Zhu, Shabnam Behzad, Amir Zeldes

We present a freely available, genre-balanced English web corpus totaling 4M tokens and featuring a large number of high-quality automatic annotation layers, including dependency trees, non-named entity annotations, coreference resolution, and discourse trees in Rhetorical Structure Theory.

coreference-resolution Coreference Resolution

A Cross-Genre Ensemble Approach to Robust Reddit Part of Speech Tagging

1 code implementation • LREC 2020 • Shabnam Behzad, Amir Zeldes

However, when these models are applied to other corpora with different genres, and especially user-generated data from the Web, we see substantial drops in performance.

Part-Of-Speech Tagging