Lune

EMNLP Findings2020

A Semi-supervised Approach to Generate the Code-Mixed Text using Pre-trained Encoder and Transfer Learning

Deepak Gupta, Asif Ekbal, Pushpak Bhattacharyya

Abstract

Code-mixing, the interleaving of two or more languages within a sentence or discourse is ubiquitous in multilingual societies. The lack of code-mixed training data is one of the major concerns for the development of end-to-end neural network-based models to be deployed for a variety of natural language processing (NLP) applications. A potential solution is to either manually create or crowd-source the code-mixed labelled data for the task at hand, but that requires much human efforts and often not feasible because of the language specific diversity in the code-mixed text. To circumvent the data scarcity issue, we propose an effective deep learning approach for automatically generating the code-mixed text from English to multiple languages without any parallel data. In order to train the neural network, we create synthetic code-mixed texts from the available parallel corpus by modelling various linguistic properties of code-mixing. Our codemixed text generator is built upon the encoderdecoder framework, where the encoder is augmented with the linguistic and task-agnostic features obtained from the transformer based language model. We also transfer the knowledge from a neural machine translation (NMT) to warm-start the training of code-mixed generator. Experimental results and in-depth analysis show the effectiveness of our proposed code-mixed text generation on eight diverse language pairs. Language (L1) Language (L2) Code-Mixed (L1-L2) en India's agriculture is their main strength. hi भारत क कृ ष उनक मु य ताकत है । India's कृ ष इसक main strength है । en Especially valuable people like Connor Rooney. bn িবেশষত কনর িনর মেতা মূ ল বান ব ি । িবেশষ Connor Rooney মেতা valuable ব ি । en Glasses and cups, whatever they are, can be turned upside down. ta க ணா க ம ேகா ைபக , அைவ எ வாக இ தா , தைல ழாக மா றலா . Glasses ம cups அைவ எ வாக இ தா , தைல ழாக மா றலா en Democracy and development go hand in hand. de Demokratie und Entwicklung gehen Hand in Hand. Democracy und Development gehen Hand in Hand. en We abolish national embassies. fr Nous abolissons les ambassades nationales. Nous abolissons les embassies national.

Ask Lune

Ask anything. The answer quotes the paper.

Lune reads all of it and the papers around it, then shows you the passages behind every answer.

Questions to start from

Lune answers from

the full paper · 1 top-tier paper that cites it · related work

Get the answer

Free to start. No credit card required.

Cited by top-tier papers1

  1. CodeMixBench: Evaluating Code-Mixing Capabilities of LLMs Across 18 LanguagesYilun Yang, Yekun ChaiEMNLP 2025 · 1 citation

Related papers

  1. Adapting Multilingual Models for Code-Mixed TranslationAditya Vavre, Abhirut Gupta, Sunita SarawagiEMNLP Findings 2022 · 4 citations
  2. CoMix: Guide Transformers to Code-Mix using POS structure and PhoneticsGaurav Arora, Srujana Merugu, Vivek SembiumACL Findings 2023
  3. From Machine Translation to Code-Switching: Generating High-Quality Code-Switched TextIshan Tarunesh, Syamantak Kumar, Preethi JyothiACL 2021
  4. SyMCoM - Syntactic Measure of Code Mixing A Study Of English-Hindi Code-MixingPrashant Kodali, Anmol Goel, Monojit Choudhury, Manish Shrivastava et al.ACL Findings 2022 · 1 citation
  5. Disentangling Codemixing in Chats: The NUS ABC Codemixed CorpusSvetlana Churina, Akshat Gupta, Nur Insyirah Binte Imam Mujtahid, Kokil JaidkaACL Findings 2026
Dusk over the sea between two cliffs drawn in fine vertical lines