Advancing NLP with Cognitive Language Processing Signals

TL;DR

This paper investigates the use of cognitive signals like gaze and EEG data to enhance NLP models across multiple tasks, demonstrating significant improvements and discussing current limitations.

Contribution

It provides an extensive analysis of how human cognitive processing data can be used to improve NLP performance across various tasks.

Findings

Gaze and EEG features improve NLP task accuracy

Models augmented with cognitive data outperform baselines

Limitations include data variability and task-specific challenges

Abstract

When we read, our brain processes language and generates cognitive processing data such as gaze patterns and brain activity. These signals can be recorded while reading. Cognitive language processing data such as eye-tracking features have shown improvements on single NLP tasks. We analyze whether using such human features can show consistent improvement across tasks and data sources. We present an extensive investigation of the benefits and limitations of using cognitive processing data for NLP. Specifically, we use gaze and EEG features to augment models of named entity recognition, relation classification, and sentiment analysis. These methods significantly outperform the baselines and show the potential and current limitations of employing human language processing data for NLP.