Abstract
This paper describes a multi-hop LLM agent for tabular question answering developed for SemEval-2025 Task 8 and ranked 6th with 87% accuracy. Our approach combines proprietary LLM (ChatGPT-3.5-turbo) for code generation and open source LLM (Llama-3.2-3B) for answer validation.- Anthology ID:
- 2025.semeval-1.174
- 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:
- 1313–1317
- Language:
- URL:
- https://preview.aclanthology.org/more-markup/2025.semeval-1.174/
- DOI:
- Cite (ACL):
- Maryna Chernyshevich. 2025. Core Intelligence at SemEval-2025 Task 8: Multi-hop LLM Agent for Tabular Question Answering. In Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025), pages 1313–1317, Vienna, Austria. Association for Computational Linguistics.
- Cite (Informal):
- Core Intelligence at SemEval-2025 Task 8: Multi-hop LLM Agent for Tabular Question Answering (Chernyshevich, SemEval 2025)
- PDF:
- https://preview.aclanthology.org/more-markup/2025.semeval-1.174.pdf