Work

UIT-Polar at SemEval-2026 Task 9 Detecting Multilingual, Multicultural and Multievent Online Polarization

Year: 2026

Type: conference-paper

Abstract: We present a two-stage hybrid system for SemEval-2026 Task 9 on multilingual and multievent online polarization detection.The first stage employs DeBERTa for high-recall binary filtering to mitigate severe class imbalance.The second stage leverages Mistral for fine-grained polarization classification, enabling improved semantic reasoning over candidate instances.This coarse-to-fine design enhances robustness and efficiency while preserving minority-class performance.UIT-Polar system achieves Top-5 results on the English test set, demonstrating the effectiveness of integrating encoder-based screening with LLM-based refinement.

Source: International Workshop on Semantic Evaluation (SemEval)

DOI: 10.18653/v1/2026.semeval-1.19

Author: Hoàn Trần

Language: English

Topic: Sentiment Analysis and Opinion Mining

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

OpenAlex ID: W7166866910

FWCI:

Cites:

Cited by:

Related to: 10