BeClaude
Research2026-04-23

COMPASS: COntinual Multilingual PEFT with Adaptive Semantic Sampling

Source: Arxiv CS.AI

arXiv:2604.20720v1 Announce Type: cross Abstract: Large language models (LLMs) often exhibit performance disparities across languages, with naive multilingual fine-tuning frequently degrading performance due to negative cross-lingual interference. To address this, we introduce COMPASS (COntinual...

arxivpapers