CALCS @ NAACL 2021

9 papers

Top Research Directions

The 28 research directions with CALCS @ NAACL 2021 papers, ranked by paper count. Share is the fraction of the edition’s papers in the direction.

#DirectionPapersShare
1Bridging the Orthographic Gap in Multilingual Transfer333.3%
2Overcoming Data Scarcity in Code-Switched Language Modeling333.3%
3Orthographic Inconsistency in Low Resource Language Processing222.2%
4Progressive Domain Adaptation via Multi-Stage Fine-Tuning222.2%
5Adapting Language Models via Continual Pre-training111.1%
6Adversarial Robustness in Generative Language Tasks111.1%
7Adversarial Training for Robust Machine Learning111.1%
8Augmenting Word Embeddings with Explicit Linguistic Features111.1%
9Breaking the Confirmation Bias Loop in Pseudo-Labeling111.1%
10Bridging the Multilingual Coverage Gap111.1%
11Bypassing the Computational Bottleneck of Adversarial Training111.1%
12Cross-Lingual Gender Mapping in Machine Translation111.1%
13Cross-Lingual Transfer via Synthetic Code-Switching111.1%
14Decoupling Encoders for Extensible Multimodal Learning111.1%
15Metrics for Quantifying Code Switching Structure111.1%
16Mitigating Monolingual Bias in Code-Switched Language Models111.1%
17Multi-Granular Representation Fusion in Language Models111.1%
18Script Normalization for Multilingual and Code Mixed Text111.1%
19Statistical Post Editing for Hybrid Machine Translation111.1%
20Subword Modeling for Intra Word Code Switching111.1%
21Synthetic Data Generation for Code-Switched Language Processing111.1%
22Text Normalization as Sequence to Sequence Translation111.1%
23The Cross-Platform Generalization Gap in Social Media NLP111.1%
24The Multilingual Evaluation Gap in Foundation Models111.1%
25The Performance Advantage of Localized Language Pretraining111.1%

Download all directions as CSV

Institutions

Institutions credited by CALCS @ NAACL 2021 papers: a paper credits every institution named in its affiliations, once. 11 institutions share 9 credited papers.

#InstitutionPapersShare
1Carnegie Mellon University 222.2%
2International Institute of Information Technology, Hyderabad 222.2%
3Guru Gobind Singh Indraprastha University 111.1%
4Hong Kong University of Science and Technology 111.1%
5Indraprastha Institute of Information Technology Delhi 111.1%
6Nanyang Technological University 111.1%
7Nara Institute of Science and Technology 111.1%
8National University of Singapore 111.1%
9Salesforce (United States) 111.1%
10University of British Columbia 111.1%
11University of Houston 111.1%

Download all institutions as CSV

About this data

Papers. CALCS @ NAACL 2021 lists 9 papers on Lacuna, duplicate versions counted once; rejected and withdrawn submissions are left out.

Directions. Directions are Lacuna’s clustering of the concept elements extracted from each paper, so a paper can belong to several and direction counts add up to more than the paper count. A direction’s share is its papers divided by the edition’s papers.

Institutions. A paper credits each institution named in its affiliations, resolved to a ROR record, once. 9 of 9 papers (100.0%) credit at least one institution and are the base for institution shares; the rest have no resolved affiliation yet.

Data. Counts come from Lacuna’s data build of 2026-11-20 and change as papers are added. Tables as CSV: directions · institutions. Cite as: Lacuna, “CALCS @ NAACL 2021”, https://lacuna.tiptreesystems.com/venue/calcs-naacl/2021.

Top Authors

AuthorPapers
Manish Shrivastava2
Akshat Gupta1
Alan Black1
Andrea Madotto1
Anmol Goel1
Chihiro Taguchi1
Genta Indra Winata1
Khyathi Chandu1
Laks V. S. Lakshmanan1
Pascale Fung1
Prashant Kodali1
Sai Muralidhar Jayanthi1
Samson Tan1
Samuel Cahyawijaya1
Sargam Menghani1
Shafiq Joty1
Taro Watanabe1
Yusuke Sakai1
Zhaojiang Lin1
Zihan Liu1

Impactful Papers

Papers from CALCS @ NAACL 2021 that shaped research clusters.

Influential

Papers

Title
Are Multilingual Models Effective in Code-Switching?
CodemixedNLP: An Extensible and Open NLP Toolkit for Code-Mixing
Code-Mixing on Sesame Street: Dawn of the Adversarial Polyglots
CoMeT: Towards Code-Mixed Translation Using Parallel Monolingual Sentences
Exploring Text-to-Text Transformers for English to Hinglish Machine Translation with Synthetic Code-Mixing
Normalization and Back-Transliteration for Code-Switched Data
Translate and Classify: Improving Sequence Level Classification for English-Hindi Code-Mixed Data
Transliteration for Low-Resource Code-Switching Texts: Building an Automatic Cyrillic-to-Latin Converter for Tatar
Unsupervised Self-Training for Sentiment Analysis of Code-Switched Data