AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

Improving Low-Resource Translation with Dictionary-Guided Fine-Tuning and RL: A Spanish-to-Wayuunaiki Study

Manuel Mosquera, Melissa Verónica Robles, Johan R. Portela, Rubén Manrique

AAAI 2026 · Large language models › Instruction and fine-tuning · approach: Efficient LLM fine-tuning systems

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add SFT with synthetic dictionary-call examplesQwen2.5-0.5B-Instruct · Spanish-Wayuunaiki testavg BLEU: 0.06 -> 3.08Improved
add GRPO RL stage with BLEU reward after SFT (ablation)Qwen2.5-0.5B-Instruct · Spanish-Wayuunaiki testavg BLEU: 3.08 -> 3.42Improved
give model access to dictionary tool (ablation)Qwen2.5-0.5B-Instruct · Spanish-Wayuunaiki testavg BLEU: about 6% relative gainImproved
use CharacTer reward instead of BLEU in RL (ablation)Qwen2.5-0.5B-Instruct · Spanish-Wayuunaiki testavg BLEU: 3.42 -> 2.76Hurt
train RL for 1400 steps instead of 400 (BLEU reward) (ablation)Qwen2.5-0.5B-Instruct · Spanish-Wayuunaiki testavg BLEU: 3.16 -> 3.42Improved
scale from Qwen2.5-0.5B to Qwen2.5-7B (ablation)Qwen2.5 · Spanish-Wayuunaiki testavg BLEU: 3.42 -> 4.45Improved
apply RL to NLLB-3B translation model (ablation)NLLB-3B · Spanish-Wayuunaiki testavg BLEU: 0.93 -> 0.93No clear effect
apply the method to Llama-3.2-1B-Instruct (ablation)Llama-3.2-1B-Instruct · Spanish-Wayuunaiki testavg BLEU: 3.15 -> 3.48 (SFT to RL)Improved
tell base model a dictionary exists without examples (ablation)Qwen2.5-0.5B-Instruct · Spanish-Wayuunaiki testavg BLEU: 0.83 -> 0.06Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy