Beyond Benchmark: LLMs Evaluation with an Anthropomorphic and Value-oriented Roadmap

TL;DR

This paper proposes a holistic, anthropomorphic evaluation framework for Large Language Models that assesses their general intelligence, emotional alignment, and professional expertise, addressing limitations of current benchmark-centric methods.

Contribution

It introduces a novel three-dimensional taxonomy and a value-oriented evaluation framework to better measure LLMs' real-world utility and ethical alignment.

Findings

Analysis of 200+ benchmarks highlights key evaluation challenges.

The proposed framework offers a comprehensive assessment of LLM capabilities.

A curated repository of open-source evaluation resources is provided.

Abstract

For Large Language Models (LLMs), a disconnect persists between benchmark performance and real-world utility. Current evaluation frameworks remain fragmented, prioritizing technical metrics while neglecting holistic assessment for deployment. This survey introduces an anthropomorphic evaluation paradigm through the lens of human intelligence, proposing a novel three-dimensional taxonomy: Intelligence Quotient (IQ)-General Intelligence for foundational capacity, Emotional Quotient (EQ)-Alignment Ability for value-based interactions, and Professional Quotient (PQ)-Professional Expertise for specialized proficiency. For practical value, we pioneer a Value-oriented Evaluation (VQ) framework assessing economic viability, social impact, ethical alignment, and environmental sustainability. Our modular architecture integrates six components with an implementation roadmap. Through analysis of 200+ benchmarks, we identify key challenges including dynamic assessment needs and interpretability gaps. It provides actionable guidance for developing LLMs that are technically proficient, contextually relevant, and ethically sound. We maintain a curated repository of open-source evaluation resources at: https://github.com/onejune2018/Awesome-LLM-Eval.