Pre-Trained Language Models Augmented with Synthetic Scanpaths for Natural Language Understanding

TL;DR

This paper introduces a method to generate synthetic gaze data to enhance language models for understanding tasks, eliminating the need for scarce human gaze data and achieving comparable performance.

Contribution

The authors develop a model that integrates synthetic scanpath generation with language models, enabling fine-tuning for NLP tasks without relying on real gaze data.

Findings

Synthetic gaze data improves language model performance.

The model outperforms baseline language models.

Performance is comparable to models using real gaze data.

Abstract

Human gaze data offer cognitive information that reflects natural language comprehension. Indeed, augmenting language models with human scanpaths has proven beneficial for a range of NLP tasks, including language understanding. However, the applicability of this approach is hampered because the abundance of text corpora is contrasted by a scarcity of gaze data. Although models for the generation of human-like scanpaths during reading have been developed, the potential of synthetic gaze data across NLP tasks remains largely unexplored. We develop a model that integrates synthetic scanpath generation with a scanpath-augmented language model, eliminating the need for human gaze data. Since the model's error gradient can be propagated throughout all parts of the model, the scanpath generator can be fine-tuned to downstream tasks. We find that the proposed model not only outperforms the underlying language model, but achieves a performance that is comparable to a language model augmented with real human gaze data. Our code is publicly available.