AILS-NTUA at SemEval-2026 Task 3: Efficient Dimensional Aspect-Based Sentiment Analysis

TL;DR

This paper introduces an efficient, multilingual system for Dimensional Aspect-Based Sentiment Analysis that combines encoder fine-tuning and instruction tuning of large language models, achieving strong, competitive results.

Contribution

The paper presents a unified, parameter-efficient approach for multiple DimABSA tasks across languages and domains, improving performance over baselines.

Findings

Achieves competitive performance across tasks and languages.

Reduces training and inference requirements.

Consistently surpasses baseline models.

Abstract

In this paper, we present AILS-NTUA system for Track-A of SemEval-2026 Task 3 on Dimensional Aspect-Based Sentiment Analysis (DimABSA), which encompasses three complementary problems: Dimensional Aspect Sentiment Regression (DimASR), Dimensional Aspect Sentiment Triplet Extraction (DimASTE), and Dimensional Aspect Sentiment Quadruplet Prediction (DimASQP) within a multilingual and multi-domain framework. Our methodology combines fine-tuning of language-appropriate encoder backbones for continuous aspect-level sentiment prediction with language-specific instruction tuning of large language models using LoRA for structured triplet and quadruplet extraction. This unified yet task-adaptive design emphasizes parameter-efficient specialization across languages and domains, enabling reduced training and inference requirements while maintaining strong effectiveness. Empirical results demonstrate that the proposed models achieve competitive performance and consistently surpass the provided baselines across most evaluation settings.