UnifiedABSA: A Unified ABSA Framework Based on Multi-task Instruction Tuning

TL;DR

UnifiedABSA introduces a multi-task instruction tuning framework that models various aspect-based sentiment analysis tasks simultaneously, improving performance and data efficiency over traditional task-specific models.

Contribution

It proposes a unified ABSA framework using multi-task instruction tuning to model diverse tasks and their inter-dependencies within a single system.

Findings

Outperforms dedicated models on 11 ABSA tasks

Demonstrates superior data efficiency

Effective in modeling inter-task dependencies

Abstract

Aspect-Based Sentiment Analysis (ABSA) aims to provide fine-grained aspect-level sentiment information. There are many ABSA tasks, and the current dominant paradigm is to train task-specific models for each task. However, application scenarios of ABSA tasks are often diverse. This solution usually requires a large amount of labeled data from each task to perform excellently. These dedicated models are separately trained and separately predicted, ignoring the relationship between tasks. To tackle these issues, we present UnifiedABSA, a general-purpose ABSA framework based on multi-task instruction tuning, which can uniformly model various tasks and capture the inter-task dependency with multi-task learning. Extensive experiments on two benchmark datasets show that UnifiedABSA can significantly outperform dedicated models on 11 ABSA tasks and show its superiority in terms of data efficiency.