Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Patidar, Mayur, Sawhney, Riya, Singh, Avinash, Chatterjee, Biswajit, Mausam, Bhattacharya, Indrajit
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2311.08894
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917691943550976
author Patidar, Mayur
Sawhney, Riya
Singh, Avinash
Chatterjee, Biswajit
Mausam
Bhattacharya, Indrajit
author_facet Patidar, Mayur
Sawhney, Riya
Singh, Avinash
Chatterjee, Biswajit
Mausam
Bhattacharya, Indrajit
contents Existing Knowledge Base Question Answering (KBQA) architectures are hungry for annotated data, which make them costly and time-consuming to deploy. We introduce the problem of few-shot transfer learning for KBQA, where the target domain offers only a few labeled examples, but a large labeled training dataset is available in a source domain. We propose a novel KBQA architecture called FuSIC-KBQA that performs KB-retrieval using multiple source-trained retrievers, re-ranks using an LLM and uses this as input for LLM few-shot in-context learning to generate logical forms. These are further refined using execution-guided feedback. Experiments over multiple source-target KBQA pairs of varying complexity show that FuSIC-KBQA significantly outperforms adaptations of SoTA KBQA models for this setting. Additional experiments show that FuSIC-KBQA also outperforms SoTA KBQA models in the in-domain setting when training data is limited.
format Preprint
id arxiv_https___arxiv_org_abs_2311_08894
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Few-shot Transfer Learning for Knowledge Base Question Answering: Fusing Supervised Models with In-Context Learning
Patidar, Mayur
Sawhney, Riya
Singh, Avinash
Chatterjee, Biswajit
Mausam
Bhattacharya, Indrajit
Computation and Language
Artificial Intelligence
Existing Knowledge Base Question Answering (KBQA) architectures are hungry for annotated data, which make them costly and time-consuming to deploy. We introduce the problem of few-shot transfer learning for KBQA, where the target domain offers only a few labeled examples, but a large labeled training dataset is available in a source domain. We propose a novel KBQA architecture called FuSIC-KBQA that performs KB-retrieval using multiple source-trained retrievers, re-ranks using an LLM and uses this as input for LLM few-shot in-context learning to generate logical forms. These are further refined using execution-guided feedback. Experiments over multiple source-target KBQA pairs of varying complexity show that FuSIC-KBQA significantly outperforms adaptations of SoTA KBQA models for this setting. Additional experiments show that FuSIC-KBQA also outperforms SoTA KBQA models in the in-domain setting when training data is limited.
title Few-shot Transfer Learning for Knowledge Base Question Answering: Fusing Supervised Models with In-Context Learning
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2311.08894