Saved in:
Bibliographic Details
Main Authors: Zolkepli, Husein, Razak, Aisyah, Adha, Kamarul, Nazhan, Ariff
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.03053
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914666960125952
author Zolkepli, Husein
Razak, Aisyah
Adha, Kamarul
Nazhan, Ariff
author_facet Zolkepli, Husein
Razak, Aisyah
Adha, Kamarul
Nazhan, Ariff
contents In this work, we present a comprehensive exploration of finetuning Malaysian language models, specifically Llama2 and Mistral, on embedding tasks involving negative and positive pairs. We release two distinct models tailored for Semantic Similarity and Retrieval-Augmented Generation (RAG). For Semantic Similarity, our 600 million parameter Llama2 model outperforms OpenAI text-embedding-ada-002 across all recall@k metrics for b.cari.com.my, c.cari.com.my, Malay news, and Malaysian Twitter test sets. In the realm of RAG models, our approach proves competitive with OpenAI text-embedding-ada-002 in the Malaysian context. Notably, our 2 billion parameter Llama2 model achieves superior Recall@5, Recall@10 for the "Melayu" keyword research papers dataset and excels in Recall@3, Recall@5, and Recall@10 for the lom.agc.gov.my dataset. These findings underscore the effectiveness of our finetuning strategy and highlight the performance gains in both Semantic Similarity and RAG tasks. All models released at https://huggingface.co/collections/mesolitica/malaysian-embedding-6523612bfe5881ad35f81b99
format Preprint
id arxiv_https___arxiv_org_abs_2402_03053
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-Lingual Malaysian Embedding: Leveraging Large Language Models for Semantic Representations
Zolkepli, Husein
Razak, Aisyah
Adha, Kamarul
Nazhan, Ariff
Computation and Language
Machine Learning
In this work, we present a comprehensive exploration of finetuning Malaysian language models, specifically Llama2 and Mistral, on embedding tasks involving negative and positive pairs. We release two distinct models tailored for Semantic Similarity and Retrieval-Augmented Generation (RAG). For Semantic Similarity, our 600 million parameter Llama2 model outperforms OpenAI text-embedding-ada-002 across all recall@k metrics for b.cari.com.my, c.cari.com.my, Malay news, and Malaysian Twitter test sets. In the realm of RAG models, our approach proves competitive with OpenAI text-embedding-ada-002 in the Malaysian context. Notably, our 2 billion parameter Llama2 model achieves superior Recall@5, Recall@10 for the "Melayu" keyword research papers dataset and excels in Recall@3, Recall@5, and Recall@10 for the lom.agc.gov.my dataset. These findings underscore the effectiveness of our finetuning strategy and highlight the performance gains in both Semantic Similarity and RAG tasks. All models released at https://huggingface.co/collections/mesolitica/malaysian-embedding-6523612bfe5881ad35f81b99
title Multi-Lingual Malaysian Embedding: Leveraging Large Language Models for Semantic Representations
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2402.03053