Modeling Comparative Logical Relation with Contrastive Learning for Text Generation

TL;DR

This paper introduces a new task in data-to-text generation focused on producing descriptions that capture deep comparative logical relations, using contrastive learning to improve understanding and generation.

Contribution

It proposes a novel contrastive learning-based method for modeling comparative logical relations in text generation and constructs a new Chinese dataset for this task.

Findings

The method outperforms baselines in automatic evaluations.

Human evaluations confirm improved logical relation accuracy.

Constructed a high-quality Chinese dataset for comparative logical relation generation.

Abstract

Data-to-Text Generation (D2T), a classic natural language generation problem, aims at producing fluent descriptions for structured input data, such as a table. Existing D2T works mainly focus on describing the superficial associative relations among entities, while ignoring the deep comparative logical relations, such as A is better than B in a certain aspect with a corresponding opinion, which is quite common in our daily life. In this paper, we introduce a new D2T task named comparative logical relation generation (CLRG). Additionally, we propose a Comparative Logic (CoLo) based text generation method, which generates texts following specific comparative logical relations with contrastive learning. Specifically, we first construct various positive and negative samples by fine-grained perturbations in entities, aspects and opinions. Then, we perform contrastive learning in the encoder layer to have a better understanding of the comparative logical relations, and integrate it in the decoder layer to guide the model to correctly generate the relations. Noting the data scarcity problem, we construct a Chinese Comparative Logical Relation Dataset (CLRD), which is a high-quality human-annotated dataset and challenging for text generation with descriptions of multiple entities and annotations on their comparative logical relations. Extensive experiments show that our method achieves impressive performance in both automatic and human evaluations.