TensorX
返回文献探索

Paper · arXiv 2401.02987

Has Your Pretrained Model Improved? A Multi-head Posterior Based Approach

Prince Aboagye, Yan Zheng, Junpeng Wang, Uday Singh Saini, Xin Dai, Michael Yeh, Yujie Fan, Zhongfang Zhuang, Shubham Jain, Liang Wang, Wei Zhang

10 upvotesJanuary 2, 2024arXiv 预印本
AI 摘要

A new evaluation method for pretrained models uses the consistency between entity representations and meta features across various domains.

pretrained modelsnatural language processingcomputer visionrelational datasetsfine-tuningmeta featuresentity representationsconsistency

Abstract

The emergence of pretrained models has significantly impacted from Natural Language Processing (NLP) and Computer Vision to relational datasets. Traditionally, these models are assessed through fine-tuned downstream tasks. However, this raises the question of how to evaluate these models more efficiently and more effectively. In this study, we explore a novel approach where we leverage the meta features associated with each entity as a source of worldly knowledge and employ entity representations from the models. We propose using the consistency between these representations and the meta features as a metric for evaluating pretrained models. Our method's effectiveness is demonstrated across various domains, including models with relational datasets, large language models and images models.

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号
Has Your Pretrained Model Improved? A Multi-head Posterior Based Approach | TensorX