REGMAPR - Text Matching Made Easy

TL;DR

REGMAPR is a simple, attention-free neural architecture for text matching that enhances Siamese models with feature augmentation, achieving state-of-the-art results on key NLP benchmarks.

Contribution

It introduces a novel, straightforward approach that improves text matching performance without relying on inter-sentence attention mechanisms.

Findings

Achieves state-of-the-art results on SICK paraphrase detection.

Performs comparably or better than complex models on multiple datasets.

Effective with regularization on various NLP tasks.

Abstract

Text matching is a fundamental problem in natural language processing. Neural models using bidirectional LSTMs for sentence encoding and inter-sentence attention mechanisms perform remarkably well on several benchmark datasets. We propose REGMAPR - a simple and general architecture for text matching that does not use inter-sentence attention. Starting from a Siamese architecture, we augment the embeddings of the words with two features based on exact and para- phrase match between words in the two sentences. We train the model using three types of regularization on datasets for textual entailment, paraphrase detection and semantic related- ness. REGMAPR performs comparably or better than more complex neural models or models using a large number of handcrafted features. REGMAPR achieves state-of-the-art results for paraphrase detection on the SICK dataset and for textual entailment on the SNLI dataset among models that do not use inter-sentence attention.