SmolLM3-3B-SFT-GRPO-ES

SmolLM3-3B-SFT-GRPO-ES

This model is the result of the second stage of the ReasonXL two-stage reasoning adaptation pipeline applied to HuggingFaceTB/SmolLM3-3B.

Stage 1 — SFT

The first stage shifts the model's reasoning language from English to Spanish through supervised fine-tuning on reasoning traces from toroe/ReasonXL-SFT.

The corresponding SFT model is:

DGurgurov/SmolLM3-3B-SFT-ES

Stage 2 — RL

This model applies RL (Dr. GRPO) to the corresponding SFT model.

The objective is to recover reasoning quality lost during supervised fine-tuning while preserving compliance with the target reasoning language.

Training uses a composite reward over verifiable mathematical problems.

Model Details

• Base model: HuggingFaceTB/SmolLM3-3B
• Target reasoning language: Spanish
• SFT dataset: toroe/ReasonXL-SFT
• Training pipeline: SFT → Dr. GRPO
• SFT model: DGurgurov/SmolLM3-3B-SFT-ES
• This model: GRPO

Full training details, reward formulation, evaluation results, and methodology will follow soon.

Citation

If you use this model, please cite:

@misc{gurgurov2026reasonxlshiftingllmreasoning,
      title={ReasonXL: Shifting LLM Reasoning Language Without Sacrificing Performance},
      author={Daniil Gurgurov and Tom Röhr and Sebastian von Rohrscheidt and Josef van Genabith and Alexander Löser and Simon Ostermann},
      year={2026},
      eprint={2604.12378},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2604.12378},
}