Saved in:
Bibliographic Details
Main Authors: Sanniboina, Saikrishna, Trivedi, Shiv, Vijayaraghavan, Sreenidhi
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.10042
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916437504819200
author Sanniboina, Saikrishna
Trivedi, Shiv
Vijayaraghavan, Sreenidhi
author_facet Sanniboina, Saikrishna
Trivedi, Shiv
Vijayaraghavan, Sreenidhi
contents Retrieval-based question answering systems often suffer from positional bias, leading to suboptimal answer generation. We propose LoRE (Logit-Ranked Retriever Ensemble), a novel approach that improves answer accuracy and relevance by mitigating positional bias. LoRE employs an ensemble of diverse retrievers, such as BM25 and sentence transformers with FAISS indexing. A key innovation is a logit-based answer ranking algorithm that combines the logit scores from a large language model (LLM), with the retrieval ranks of the passages. Experimental results on NarrativeQA, SQuAD demonstrate that LoRE significantly outperforms existing retrieval-based methods in terms of exact match and F1 scores. On SQuAD, LoRE achieves 14.5\%, 22.83\%, and 14.95\% improvements over the baselines for ROUGE-L, EM, and F1, respectively. Qualitatively, LoRE generates more relevant and accurate answers, especially for complex queries.
format Preprint
id arxiv_https___arxiv_org_abs_2410_10042
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LoRE: Logit-Ranked Retriever Ensemble for Enhancing Open-Domain Question Answering
Sanniboina, Saikrishna
Trivedi, Shiv
Vijayaraghavan, Sreenidhi
Computation and Language
Retrieval-based question answering systems often suffer from positional bias, leading to suboptimal answer generation. We propose LoRE (Logit-Ranked Retriever Ensemble), a novel approach that improves answer accuracy and relevance by mitigating positional bias. LoRE employs an ensemble of diverse retrievers, such as BM25 and sentence transformers with FAISS indexing. A key innovation is a logit-based answer ranking algorithm that combines the logit scores from a large language model (LLM), with the retrieval ranks of the passages. Experimental results on NarrativeQA, SQuAD demonstrate that LoRE significantly outperforms existing retrieval-based methods in terms of exact match and F1 scores. On SQuAD, LoRE achieves 14.5\%, 22.83\%, and 14.95\% improvements over the baselines for ROUGE-L, EM, and F1, respectively. Qualitatively, LoRE generates more relevant and accurate answers, especially for complex queries.
title LoRE: Logit-Ranked Retriever Ensemble for Enhancing Open-Domain Question Answering
topic Computation and Language
url https://arxiv.org/abs/2410.10042