ProxyLM: predicting language model performance on multilingual tasks via proxy models

Performance prediction is a method to estimate the performance of multilingual language models (LMs), mitigating computational costs associated with model capacity and data for fine-tuning. Our paper introduces ProxyLM, a scalable framework for predicting LM performance using proxy models in multilingual tasks. These proxy models act as surrogates, approximating the performance of fine-tuned LMs on specific downstream natural language processing (NLP) tasks. By leveraging proxy models, ProxyLM significantly reduces computational overhead on task evaluations, achieving up to a 37.08x speedup compared to traditional methods, even with our smallest proxy models. Additionally, our methodology showcases adaptability to previously unseen languages in pre-trained LMs, outperforming the state-of-the-art performance by 1.89x as measured by root-mean-square-error (RMSE). This framework streamlines model selection, enabling efficient deployment and iterative LM enhancements without extensive computational resources.