Evolution without an Oracle: Driving Effective Evolution with LLM Judges

TL;DR

This paper introduces MADE, a framework that enables evolution driven solely by subjective LLM judges, overcoming the need for objective fitness functions and achieving significant improvements in complex benchmarks.

Contribution

MADE decomposes vague instructions into verifiable sub-requirements, allowing effective evolution with subjective LLM feedback, a novel approach in evolutionary computation.

Findings

MADE outperforms baselines by over 50% in requirement satisfaction

Achieves 95% pass rate on instruction following

Validates evolution without objective fitness functions

Abstract

The integration of Large Language Models (LLMs) with Evolutionary Computation (EC) has unlocked new frontiers in scientific discovery but remains shackled by a fundamental constraint: the reliance on an Oracle--an objective, machine-computable fitness function. This paper breaks this barrier by asking: Can evolution thrive in a purely subjective landscape governed solely by LLM judges? We introduce MADE (Multi-Agent Decomposed Evolution), a framework that tames the inherent noise of subjective evaluation through "Problem Specification." By decomposing vague instructions into specific, verifiable sub-requirements, MADE transforms high-variance LLM feedback into stable, precise selection pressure. The results are transformative: across complex benchmarks like DevAI and InfoBench, MADE outperforms strong baselines by over 50% in software requirement satisfaction (39.9% to 61.9%) and achieves a 95% perfect pass rate on complex instruction following. This work validates a fundamental paradigm shift: moving from optimizing "computable metrics" to "describable qualities," thereby unlocking evolutionary optimization for the vast open-ended domains where no ground truth exists.