@inproceedings{yao-yang-2026-xiaom,
    title = "{X}iao{M} at {S}em{E}val-2026 Task 7: A Qwen-based System for Accurate Retrieval of Everyday Knowledge Across Diverse Languages and Cultures",
    author = "Yao, Xiao  and
      Yang, Liang",
    editor = "Kochmar, Ekaterina  and
      Ghosh, Debanjan  and
      North, Kai  and
      Komachi, Mamoru",
    booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.semeval-1.213/",
    pages = "1659--1663",
    ISBN = "979-8-89176-414-9",
    abstract = "This paper describes our system designed for SemEval-2026 Task 7: Everyday Knowledge Across Diverse Languages and Cultures. We describe a practical inference system for a two-track benchmark consisting of short-answer questions (SAQ) and multiple-choice questions (MCQ). Our submission is implemented in a single script and targets competition constraints directly: strict TSV schemas, short answer limits, and reliability under batch inference. The system uses Qwen2.5-7B-Instruct with memory-aware initialization, deterministic decoding (no sampling, zero temperature), and post-processing rules that guarantee valid outputs. We further add retry-on-failure and file-write fault tolerance to reduce runtime interruptions."
}