# Assessing incrementality in sequence-to-sequence models

## TL;DR

This paper evaluates how sequence-to-sequence models process sentences incrementally, comparing attention-based and non-attention models, and introduces metrics to analyze their behavior in relation to human language processing.

## Contribution

It proposes three novel metrics to assess the incremental processing behavior of RNNs with and without attention mechanisms, highlighting key differences.

## Key findings

- Attention mechanisms may hinder incremental processing.
- RNNs without attention process sentences more incrementally.
- Key differences identified between model types in sentence processing.

## Abstract

Since their inception, encoder-decoder models have successfully been applied to a wide array of problems in computational linguistics. The most recent successes are predominantly due to the use of different variations of attention mechanisms, but their cognitive plausibility is questionable. In particular, because past representations can be revisited at any point in time, attention-centric methods seem to lack an incentive to build up incrementally more informative representations of incoming sentences. This way of processing stands in stark contrast with the way in which humans are believed to process language: continuously and rapidly integrating new information as it is encountered. In this work, we propose three novel metrics to assess the behavior of RNNs with and without an attention mechanism and identify key differences in the way the different model types process sentences.