Diff-eRank: A Novel Rank-Based Metric for Evaluating Large Language Models

Lai Wei, Zhiquan Tan, Chenghai Li, Jindong Wang, Weiran Huang

Keywords: Large Language Model, Multi-modal Large Language Model, Evaluation Metric

TL;DR: We propose a novel rank-based metric, Diff-eRank, to evaluate LLMs based on representations.

Abstract:

Large Language Models (LLMs) have transformed natural language processing and extended their powerful capabilities to multi-modal domains. As LLMs continue to advance, it is crucial to develop diverse and appropriate metrics for their evaluation. In this paper, we introduce a novel rank-based metric, Diff-eRank, grounded in information theory and geometry principles. Diff-eRank assesses LLMs by analyzing their hidden representations, providing a quantitative measure of how efficiently they eliminate redundant information during training. We demonstrate the applicability of Diff-eRank in both single-modal (e.g., language) and multi-modal settings. For language models, our results show that Diff-eRank increases with model size and correlates well with conventional metrics such as loss and accuracy. In the multi-modal context, we propose an alignment evaluation method based on the eRank, and verify that contemporary multi-modal LLMs exhibit strong alignment performance based on our method. Our code is publicly available at https://github.com/waltonfuture/Diff-eRank.

Primary Area: Evaluation (methodology, meta studies, replicability and validity)

Submission Number: 3360