Anveshana: A New Benchmark Dataset for Cross-Lingual Information Retrieval On English Queries and Sanskrit Documents

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jagadeeshan, Manoj Balaji, Raj, Prince, Goyal, Pawan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909623049519104
author Jagadeeshan, Manoj Balaji
Raj, Prince
Goyal, Pawan
author_facet Jagadeeshan, Manoj Balaji
Raj, Prince
Goyal, Pawan
contents The study presents a comprehensive benchmark for retrieving Sanskrit documents using English queries, focusing on the chapters of the Srimadbhagavatam. It employs a tripartite approach: Direct Retrieval (DR), Translation-based Retrieval (DT), and Query Translation (QT), utilizing shared embedding spaces and advanced translation methods to enhance retrieval systems in a RAG framework. The study fine-tunes state-of-the-art models for Sanskrit's linguistic nuances, evaluating models such as BM25, REPLUG, mDPR, ColBERT, Contriever, and GPT-2. It adapts summarization techniques for Sanskrit documents to improve QA processing. Evaluation shows DT methods outperform DR and QT in handling the cross-lingual challenges of ancient texts, improving accessibility and understanding. A dataset of 3,400 English-Sanskrit query-document pairs underpins the study, aiming to preserve Sanskrit scriptures and share their philosophical importance widely. Our dataset is publicly available at https://huggingface.co/datasets/manojbalaji1/anveshana
format Preprint
id arxiv_https___arxiv_org_abs_2505_19494
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Anveshana: A New Benchmark Dataset for Cross-Lingual Information Retrieval On English Queries and Sanskrit Documents
Jagadeeshan, Manoj Balaji
Raj, Prince
Goyal, Pawan
Computation and Language
Information Retrieval
The study presents a comprehensive benchmark for retrieving Sanskrit documents using English queries, focusing on the chapters of the Srimadbhagavatam. It employs a tripartite approach: Direct Retrieval (DR), Translation-based Retrieval (DT), and Query Translation (QT), utilizing shared embedding spaces and advanced translation methods to enhance retrieval systems in a RAG framework. The study fine-tunes state-of-the-art models for Sanskrit's linguistic nuances, evaluating models such as BM25, REPLUG, mDPR, ColBERT, Contriever, and GPT-2. It adapts summarization techniques for Sanskrit documents to improve QA processing. Evaluation shows DT methods outperform DR and QT in handling the cross-lingual challenges of ancient texts, improving accessibility and understanding. A dataset of 3,400 English-Sanskrit query-document pairs underpins the study, aiming to preserve Sanskrit scriptures and share their philosophical importance widely. Our dataset is publicly available at https://huggingface.co/datasets/manojbalaji1/anveshana
title Anveshana: A New Benchmark Dataset for Cross-Lingual Information Retrieval On English Queries and Sanskrit Documents
topic Computation and Language
Information Retrieval
url https://arxiv.org/abs/2505.19494