Work

YNU-HPCC at SemEval-2026 Task 11: Mitigating Content Effects in Syllogistic Reasoning with Qwen2-1.5B-Instruct and XLM-RoBERTa-Large for English and Multilingual TasksMultilingual Tasks

Year: 2026

Type: conference-paper

Abstract: This paper addresses SemEval-2026 Task 11, which focused on mitigating content effects in syllogistic reasoning.Logical validity is often conflated with semantic plausibility in large language models.Prior methods rely on standard fine-tuning or prompting, without explicit bias control.A ruleand template-based symbolic data augmentation framework is proposed for fine-tuning the Qwen2-1.5B-Instructmodel and instructiontuning the XLM-RoBERTa-large model.Logicpreserving synthetic data are generated through lexical rules.The system is ranked 1st in Task 1 with a perfect overall score of 100, and 6th in Task 3 with a score of 56.97.Code is publicly available at: https://github.com/ YNU-HPCC/semeval-2026-task11.

Source: International Workshop on Semantic Evaluation (SemEval)

DOI: 10.18653/v1/2026.semeval-1.40

Authors: Rongchuan Luo, Jin Wang, Xuejie Zhang

Institution: Yunnan University

Language: English

Topic: Topic Modeling

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: National Natural Science Foundation of China, Yunnan University, Yunnan Provincial Department of Education

Awards: 基于多维度连续情感空间的文本生成技术研究, G5917579187