analiza aiqbench
This empirical study challenges the assumption that English serves as a universal optimal source language for cross-lingual few-shot learning in large models. By analyzing seven tasks across six architectures, researchers demonstrate that fine-tuning-based heuristics often fail to translate directly to In-Context Learning regimes due to distinct mechanisms like language confusion. The findings provide actionable alternatives for selecting high-performing source languages when deploying multilingual AI systems with limited data.
When English Isn't the Best Teacher: Source Language Effects in Cross-Lingual In-Context Learning
arXiv:2606.18033v1 Announce Type: cross Abstract: Cross-lingual transfer in multilingual NLP has been widely explored in supervised fine-tuning contexts, where factors like data availability and linguistic similarity largely determine transfer quality. As the field shifts toward few-shot In-Context Learning (ICL), it is often presumed that insights from fine-tuning carry over unchanged. Yet this assumption has not been rigorously evaluated, leaving open the question of how to choose source languages for cross-lingual ICL. We conduct a broad empirical study of cross-lingual transfer in ICL spanning seven tasks, six models, and a typologically diverse set of languages. We further analyze language confusion, a key obstacle for generative tasks in cross-lingual ICL. Our results show that conventional fine-tuning-based expectations do not consistently apply in the ICL regime and point to alternative heuristics for selecting source languages effectively.
Abstract:Cross-lingual transfer in multilingual NLP has been widely explored in supervised fine-tuning contexts, where factors like data availability and linguistic similarity largely determine transfer quality. As the field shifts toward few-shot In-Context Learning (ICL), it is often presumed that insights from fine-tuning carry over unchanged. Yet this assumption has not been rigorously evaluated, leaving open the question of how to choose source languages for cross-lingual ICL. We conduct a broad empirical study of cross-lingual transfer in ICL spanning seven tasks, six models, and a typologically diverse set of languages. We further analyze language confusion, a key obstacle for generative tasks in cross-lingual ICL. Our results show that conventional fine-tuning-based expectations do not consistently apply in the ICL regime and point to alternative heuristics for selecting source languages effectively.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2606.18033 [cs.CL]
(or arXiv:2606.18033v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2606.18033 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 226 cuvinte
Subiecte conexe
Links #3: 2026/06 Part 1
LessWrong · 15 iun. 2026
Granite Embedding Multilingual R2: Open Apache 2.0 Multilingual Embeddings with 32K Context — Best Sub-100M Retrieval Quality
Hugging Face · 14 mai 2026
[AINews] The Future of Latent Space
Latent Space · 25 sept. 2026
[AINews] Opus 5.5 is good at explainer videos
Latent Space · 29 sept. 2026