Abstract
In this paper the winning solution of PolEval 2025 Task 2a is presented. The approach utilizes LoRA fine-tuning of the Qwen3-8B model. Multiple LoRA matrix ranks are explored. Versions with and without the system prompt in loss calculation are evaluated. New SOTA was established at F1=0.6039 beating the previously best model at F1=0.5985. After the task’s conclusion the solution was improved upon and F1=0.6283±0.0056 was achieved.- Anthology ID:
- 2025.poleval-main.7
- Volume:
- Proceedings of the PolEval 2025 Workshop
- Month:
- November
- Year:
- 2025
- Address:
- Warsaw
- Editors:
- Łukasz Kobyliński, Alina Wróblewska, Maciej Ogrodniczuk
- Venues:
- PolEval | WS
- SIG:
- Publisher:
- Institute of Computer Science PAS and Association for Computational Linguistics
- Note:
- Pages:
- 48–53
- Language:
- URL:
- https://preview.aclanthology.org/ingest-paclic/2025.poleval-main.7/
- DOI:
- Cite (ACL):
- Adam Majczyk. 2025. Less is More—Achieving SOTA at PolEval 2025 Task 2a: Gender-inclusive LLMs for Polish (Proofreading) with LoRA and Qwen3-8B. In Proceedings of the PolEval 2025 Workshop, pages 48–53, Warsaw. Institute of Computer Science PAS and Association for Computational Linguistics.
- Cite (Informal):
- Less is More—Achieving SOTA at PolEval 2025 Task 2a: Gender-inclusive LLMs for Polish (Proofreading) with LoRA and Qwen3-8B (Majczyk, PolEval 2025)
- PDF:
- https://preview.aclanthology.org/ingest-paclic/2025.poleval-main.7.pdf