Interpretable Question Answering with Knowledge Graphs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Aneja, Kartikeya, Srivastava, Manasvi, Das, Subhayan, Aneja, Nagender
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909863357972480
author Aneja, Kartikeya
Srivastava, Manasvi
Das, Subhayan
Aneja, Nagender
author_facet Aneja, Kartikeya
Srivastava, Manasvi
Das, Subhayan
Aneja, Nagender
contents This paper presents a question answering system that operates exclusively on a knowledge graph retrieval without relying on retrieval augmented generation (RAG) with large language models (LLMs). Instead, a small paraphraser model is used to paraphrase the entity relationship edges retrieved from querying the knowledge graph. The proposed pipeline is divided into two main stages. The first stage involves pre-processing a document to generate sets of question-answer (QA) pairs. The second stage converts these QAs into a knowledge graph from which graph-based retrieval is performed using embeddings and fuzzy techniques. The graph is queried, re-ranked, and paraphrased to generate a final answer. This work includes an evaluation using LLM-as-a-judge on the CRAG benchmark, which resulted in accuracies of 71.9% and 54.4% using LLAMA-3.2 and GPT-3.5-Turbo, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2510_19181
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Interpretable Question Answering with Knowledge Graphs
Aneja, Kartikeya
Srivastava, Manasvi
Das, Subhayan
Aneja, Nagender
Computation and Language
Artificial Intelligence
Machine Learning
This paper presents a question answering system that operates exclusively on a knowledge graph retrieval without relying on retrieval augmented generation (RAG) with large language models (LLMs). Instead, a small paraphraser model is used to paraphrase the entity relationship edges retrieved from querying the knowledge graph. The proposed pipeline is divided into two main stages. The first stage involves pre-processing a document to generate sets of question-answer (QA) pairs. The second stage converts these QAs into a knowledge graph from which graph-based retrieval is performed using embeddings and fuzzy techniques. The graph is queried, re-ranked, and paraphrased to generate a final answer. This work includes an evaluation using LLM-as-a-judge on the CRAG benchmark, which resulted in accuracies of 71.9% and 54.4% using LLAMA-3.2 and GPT-3.5-Turbo, respectively.
title Interpretable Question Answering with Knowledge Graphs
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2510.19181