DIMSIM: Distilled Multilingual Critics for Indic Text Simplification

Sneha Mondal, Ritika Ritika, Ashish Agrawal, Preethi Jyothi, Aravindan Raghuveer


Abstract
Self-correction techniques have recently emerged as a promising framework to improve the quality of responses generated by large language models (LLMs). Few-shot prompted LLMs act as critics to produce feedback for an input, which is further fed to a refiner (also an LLM) to produce an output. However, these critique-refine steps require multiple expensive LLM calls. To circumvent this large inference cost, we borrow inspiration from prior work on knowledge distillation and propose the use of critique distillation to train critic models. These are smaller sequence-to-sequence models that are trained on input-critique pairs generated by an LLM. We focus on the problem of text simplification for three Indian languages: Hindi, Bengali and Marathi. This task is a good fit for self-correction style techniques. It also hasn’t been systematically explored for Indian languages before. We train two separate critics that focus on lexical and structure complexity, and show that it is surprisingly more effective than using an LLM directly as a critic in both 0-shot and few-shot settings. We also show the benefits of training multilingual critics, as opposed to monolingual critics. Extensive human evaluations show that on average, raters find 80% of DIMSIM’s output to be simple and easy to read.
Anthology ID:
2024.findings-acl.952
Volume:
Findings of the Association for Computational Linguistics: ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
16093–16109
Language:
URL:
https://aclanthology.org/2024.findings-acl.952
DOI:
10.18653/v1/2024.findings-acl.952
Bibkey:
Cite (ACL):
Sneha Mondal, Ritika Ritika, Ashish Agrawal, Preethi Jyothi, and Aravindan Raghuveer. 2024. DIMSIM: Distilled Multilingual Critics for Indic Text Simplification. In Findings of the Association for Computational Linguistics: ACL 2024, pages 16093–16109, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
DIMSIM: Distilled Multilingual Critics for Indic Text Simplification (Mondal et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-acl.952.pdf