Saved in:
Bibliographic Details
Main Authors: Tourni, Isidora Chara, Ghosh, Sayontan, Miao, Brenda, van der Poel, Constantijn
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.21501
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929564150661120
author Tourni, Isidora Chara
Ghosh, Sayontan
Miao, Brenda
van der Poel, Constantijn
author_facet Tourni, Isidora Chara
Ghosh, Sayontan
Miao, Brenda
van der Poel, Constantijn
contents This paper explores the problems of Question Answering (QA) and Named Entity Recognition (NER) in five diverse languages. We tested five Large Language Models with various prompting methods, including zero-shot, chain-of-thought reasoning, and translation techniques. Our results show that while some models consistently outperform others, their effectiveness varies significantly across tasks and languages. We saw that advanced prompting techniques generally improved QA performance but had mixed results for NER; and we observed that language difficulty patterns differed between tasks. Our findings highlight the need for task-specific approaches in multilingual NLP and suggest that current models may develop different linguistic competencies for different tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2410_21501
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SandboxAQ's submission to MRL 2024 Shared Task on Multi-lingual Multi-task Information Retrieval
Tourni, Isidora Chara
Ghosh, Sayontan
Miao, Brenda
van der Poel, Constantijn
Computation and Language
This paper explores the problems of Question Answering (QA) and Named Entity Recognition (NER) in five diverse languages. We tested five Large Language Models with various prompting methods, including zero-shot, chain-of-thought reasoning, and translation techniques. Our results show that while some models consistently outperform others, their effectiveness varies significantly across tasks and languages. We saw that advanced prompting techniques generally improved QA performance but had mixed results for NER; and we observed that language difficulty patterns differed between tasks. Our findings highlight the need for task-specific approaches in multilingual NLP and suggest that current models may develop different linguistic competencies for different tasks.
title SandboxAQ's submission to MRL 2024 Shared Task on Multi-lingual Multi-task Information Retrieval
topic Computation and Language
url https://arxiv.org/abs/2410.21501