UniSA: Unified Generative Framework for Sentiment Analysis

TL;DR

This paper introduces UniSA, a unified multimodal generative framework for sentiment analysis that effectively models various subtasks and outperforms existing methods on multiple benchmarks.

Contribution

The paper proposes a novel Task-Specific Prompt method and a unified generative framework, UniSA, to address challenges in unifying sentiment analysis subtasks and introduces SAEval benchmark.

Findings

UniSA achieves state-of-the-art performance on multiple sentiment analysis subtasks.

The framework generalizes well across different sentiment analysis tasks.

Novel pre-training tasks enhance multimodal sentiment perception.

Abstract

Sentiment analysis is a crucial task that aims to understand people's emotional states and predict emotional categories based on multimodal information. It consists of several subtasks, such as emotion recognition in conversation (ERC), aspect-based sentiment analysis (ABSA), and multimodal sentiment analysis (MSA). However, unifying all subtasks in sentiment analysis presents numerous challenges, including modality alignment, unified input/output forms, and dataset bias. To address these challenges, we propose a Task-Specific Prompt method to jointly model subtasks and introduce a multimodal generative framework called UniSA. Additionally, we organize the benchmark datasets of main subtasks into a new Sentiment Analysis Evaluation benchmark, SAEval. We design novel pre-training tasks and training methods to enable the model to learn generic sentiment knowledge among subtasks to improve the model's multimodal sentiment perception ability. Our experimental results show that UniSA performs comparably to the state-of-the-art on all subtasks and generalizes well to various subtasks in sentiment analysis.