End-to-End Entity Resolution and Question Answering Using Differentiable Knowledge Graphs

TL;DR

This paper introduces a fully differentiable end-to-end model for question answering over knowledge graphs that jointly learns entity resolution without requiring hand-annotated entities, simplifying the QA pipeline.

Contribution

It extends end-to-end learning for KGQA to include entity resolution, enabling training solely on question text and answer entities, and removes the need for external ER during inference.

Findings

Model performs close to baseline models with hand-annotated entities.

The approach is fully differentiable and trainable with weak supervision.

Evaluated on two public datasets, demonstrating competitive results.

Abstract

Recently, end-to-end (E2E) trained models for question answering over knowledge graphs (KGQA) have delivered promising results using only a weakly supervised dataset. However, these models are trained and evaluated in a setting where hand-annotated question entities are supplied to the model, leaving the important and non-trivial task of entity resolution (ER) outside the scope of E2E learning. In this work, we extend the boundaries of E2E learning for KGQA to include the training of an ER component. Our model only needs the question text and the answer entities to train, and delivers a stand-alone QA model that does not require an additional ER component to be supplied during runtime. Our approach is fully differentiable, thanks to its reliance on a recent method for building differentiable KGs (Cohen et al., 2020). We evaluate our E2E trained model on two public datasets and show that it comes close to baseline models that use hand-annotated entities.