Multilingual Text Style Transfer: Datasets & Models for Indian Languages

TL;DR

This paper introduces new datasets and evaluates models for sentiment-based text style transfer across eight Indian languages, highlighting the importance of parallel data and multilingual approaches.

Contribution

It provides the first comprehensive study of sentiment transfer in multiple Indian languages, including datasets and benchmark evaluations of various models.

Findings

Parallel data significantly improves TST performance

Masked Style Filling (MSF) is effective for non-parallel TST

Cross-lingual and multilingual models show promising results

Abstract

Text style transfer (TST) involves altering the linguistic style of a text while preserving its core content. This paper focuses on sentiment transfer, a popular TST subtask, across a spectrum of Indian languages: Hindi, Magahi, Malayalam, Marathi, Punjabi, Odia, Telugu, and Urdu, expanding upon previous work on English-Bangla sentiment transfer (Mukherjee et al., 2023). We introduce dedicated datasets of 1,000 positive and 1,000 negative style-parallel sentences for each of these eight languages. We then evaluate the performance of various benchmark models categorized into parallel, non-parallel, cross-lingual, and shared learning approaches, including the Llama2 and GPT-3.5 large language models (LLMs). Our experiments highlight the significance of parallel data in TST and demonstrate the effectiveness of the Masked Style Filling (MSF) approach (Mukherjee et al., 2023) in non-parallel techniques. Moreover, cross-lingual and joint multilingual learning methods show promise, offering insights into selecting optimal models tailored to the specific language and task requirements. To the best of our knowledge, this work represents the first comprehensive exploration of the TST task as sentiment transfer across a diverse set of languages.