Task-Routed Mixture-of-Experts with Cognitive Appraisal for Implicit Sentiment Analysis

TL;DR

This paper introduces a novel multi-task learning framework using task-specific mixture-of-experts for implicit sentiment analysis, inspired by cognitive appraisal theory, improving performance on challenging sentiment inference tasks.

Contribution

It proposes a task-routed mixture-of-experts model with a task-conditioned router and separate routing objectives, enhancing flexibility and reducing task interference in implicit sentiment analysis.

Findings

Outperforms recent approaches on implicit sentiment tasks.

Achieves significant gains on the implicit sentiment subset.

Demonstrates the effectiveness of task-specific expert routing.

Abstract

Implicit sentiment analysis is challenging because sentiment toward an aspect is often inferred from events rather than expressed through explicit opinion words. Existing models typically learn from the final polarity label, which provides limited guidance for reasoning about sentiment from the context. Motivated by cognitive appraisal theory, we propose an appraisal-aware multi-task learning (MTL) framework for implicit sentiment analysis that provides polarity prediction with two complementary auxiliary tasks: implicit sentiment detection and cognitive rationale generation. However, training several objectives with different targets and sharing a single backbone across tasks in MTL limits flexibility and can lead to task interference. To reduce interference among these related but distinct objectives, we adopt task-level mixture-of-experts models in which all tasks share a common set of experts, and task identity controls the sparse combination of these experts. Our method builds on an encoder-decoder architecture and replaces a subset of encoder and decoder blocks with these sparse mixtures. We use a task-conditioned router to select sparse expert mixtures for each task, and a task-separated routing objective to encourage different tasks to learn distinct expert-selection patterns. Experimental results show that our model outperforms recently proposed approaches, with strong gains on the implicit sentiment subset. Our code is available at https://github.com/yaping166/TRMoE-ISA.