HalluSearch at SemEval-2025 Task 3: A Search-Enhanced RAG Pipeline for Hallucination Detection

Mohamed Abdallah, Samhaa El - Beltagy


Abstract
We present HalluSearch, a multilingual pipeline designed to detect fabricated text spans in Large Language Model (LLM) outputs as part of Mu-SHROOM. HalluSearch couples retrieval-augmented verification with fine-grained factual splitting to identify and localize hallucinations in 14 different languages. Empirical evaluations show that HalluSearch performs competitively, placing fourth in both English (within the top 10%) and Czech. While the system’s retrieval-based strategy generally proves robust, it faces challenges in languages with limited online coverage, underscoring the need for further research to ensure consistent hallucination detection across diverse linguistic contexts.
Anthology ID:
2025.semeval-1.189
Volume:
Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)
Month:
July
Year:
2025
Address:
Vienna, Austria
Editors:
Sara Rosenthal, Aiala Rosá, Debanjan Ghosh, Marcos Zampieri
Venues:
SemEval | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1436–1441
Language:
URL:
https://preview.aclanthology.org/corrections-2025-08/2025.semeval-1.189/
DOI:
Bibkey:
Cite (ACL):
Mohamed Abdallah and Samhaa El - Beltagy. 2025. HalluSearch at SemEval-2025 Task 3: A Search-Enhanced RAG Pipeline for Hallucination Detection. In Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025), pages 1436–1441, Vienna, Austria. Association for Computational Linguistics.
Cite (Informal):
HalluSearch at SemEval-2025 Task 3: A Search-Enhanced RAG Pipeline for Hallucination Detection (Abdallah & El - Beltagy, SemEval 2025)
Copy Citation:
PDF:
https://preview.aclanthology.org/corrections-2025-08/2025.semeval-1.189.pdf