Reinforcement learning with semantic rewards lets LLMs gain low-resource language skills without the alignment tax that degrades general capabilities in supervised fine-tuning.
Llama beyond english: An empirical study on language capability transfer
5 Pith papers cite this work, alongside 6 external citations. Polarity classification is still indexing.
5
Pith papers citing it
6
external citations · external index
years
2026 5verdicts
UNVERDICTED 5representative citing papers
MSD enables cross-lingual safety transfer in LLMs via self-distillation with Dual-Perspective Safety Weighting, improving safety in low-resource languages without target response data.
LLMs handle basic arithmetic reliably in Sinhala and Tamil but show clear performance drops on complex math tasks compared to English.
Replacing tokens, freezing the corresponding embeddings, and tuning the rest of the model improves NLU performance on low-resource languages compared to full fine-tuning.
Analysis of open-weight LLMs reveals strong English bias in generated sequences, with continual pre-training providing no cost benefit over from-scratch training for non-English adaptation.
citing papers explorer
-
Reinforcement Learning with Semantic Rewards Enables Low-Resource Language Expansion without Alignment Tax
Reinforcement learning with semantic rewards lets LLMs gain low-resource language skills without the alignment tax that degrades general capabilities in supervised fine-tuning.
-
Multilingual Safety Alignment via Self-Distillation
MSD enables cross-lingual safety transfer in LLMs via self-distillation with Dual-Perspective Safety Weighting, improving safety in low-resource languages without target response data.
-
Large Language Models for Math Education in Low-Resource Languages: A Study in Sinhala and Tamil
LLMs handle basic arithmetic reliably in Sinhala and Tamil but show clear performance drops on complex math tasks compared to English.
-
Modular Monolingual Adaptation using Pretrained Language Models
Replacing tokens, freezing the corresponding embeddings, and tuning the rest of the model improves NLU performance on low-resource languages compared to full fine-tuning.
-
Toward LLMs Beyond English-Centric Development
Analysis of open-weight LLMs reveals strong English bias in generated sequences, with continual pre-training providing no cost benefit over from-scratch training for non-English adaptation.