# Deep Joint Entity Disambiguation with Local Neural Attention

## TL;DR

This paper introduces a deep learning model that uses neural attention and joint inference for improved document-level entity disambiguation, achieving competitive or state-of-the-art accuracy efficiently.

## Contribution

It presents a novel neural attention-based model for joint entity disambiguation that integrates deep learning with traditional probabilistic methods.

## Key findings

- Achieves state-of-the-art accuracy on entity disambiguation tasks.
- Operates with moderate computational costs.
- Demonstrates the effectiveness of neural attention in disambiguation.

## Abstract

We propose a novel deep learning model for joint document-level entity disambiguation, which leverages learned neural representations. Key components are entity embeddings, a neural attention mechanism over local context windows, and a differentiable joint inference stage for disambiguation. Our approach thereby combines benefits of deep learning with more traditional approaches such as graphical models and probabilistic mention-entity maps. Extensive experiments show that we are able to obtain competitive or state-of-the-art accuracy at moderate computational costs.