ENCONTER: Entity Constrained Progressive Sequence Generation via Insertion-based Transformer

TL;DR

ENCONTER is a novel insertion transformer model that effectively incorporates hard lexical constraints, such as entities, during sequence generation, overcoming early termination issues and improving performance in constrained text generation tasks.

Contribution

The paper introduces ENCONTER, a new insertion transformer with a training strategy for hard lexical constraints, enhancing sequence generation control without sacrificing efficiency.

Findings

ENCONTER outperforms baseline models on multiple metrics.

The model effectively incorporates predefined entities during generation.

It reduces early termination issues in constrained sequence generation.

Abstract

Pretrained using large amount of data, autoregressive language models are able to generate high quality sequences. However, these models do not perform well under hard lexical constraints as they lack fine control of content generation process. Progressive insertion-based transformers can overcome the above limitation and efficiently generate a sequence in parallel given some input tokens as constraint. These transformers however may fail to support hard lexical constraints as their generation process is more likely to terminate prematurely. The paper analyses such early termination problems and proposes the Entity-constrained insertion transformer (ENCONTER), a new insertion transformer that addresses the above pitfall without compromising much generation efficiency. We introduce a new training strategy that considers predefined hard lexical constraints (e.g., entities to be included in the generated sequence). Our experiments show that ENCONTER outperforms other baseline models in several performance metrics rendering it more suitable in practical applications. Our code is available at https://github.com/LARC-CMU-SMU/Enconter