A2Eval: Agentic and Automated Evaluation for Embodied Brain

TL;DR

A2Eval introduces an autonomous framework that streamlines embodied vision-language model evaluation by reducing costs, balancing benchmarks, and improving ranking accuracy through agentic automation.

Contribution

It is the first to automate benchmark curation and evaluation using collaborative agents, significantly reducing resource use and bias in embodied model assessment.

Findings

Compresses evaluation suites by 85%

Reduces computational costs by 77%

Achieves 4.6x speedup and high ranking fidelity

Abstract

Current embodied VLM evaluation relies on static, expert-defined, manually annotated benchmarks that exhibit severe redundancy and coverage imbalance. This labor intensive paradigm drains computational and annotation resources, inflates costs, and distorts model rankings, ultimately stifling iterative development. To address this, we propose Agentic Automatic Evaluation (A2Eval), the first agentic framework that automates benchmark curation and evaluation through two collaborative agents. The Data Agent autonomously induces capability dimensions and assembles a balanced, compact evaluation suite, while the Eval Agent synthesizes and validates executable evaluation pipelines, enabling fully autonomous, high-fidelity assessment. Evaluated across 10 benchmarks and 13 models, A2Eval compresses evaluation suites by 85%, reduces overall computational costs by 77%, and delivers a 4.6x speedup while preserving evaluation quality. Crucially, A2Eval corrects systematic ranking biases, improves human alignment to Spearman's rho=0.85, and maintains high ranking fidelity (Kendall's tau=0.81), establishing a new standard for high-fidelity, low-cost embodied assessment. Our code and data will be public soon.