DEI: Diversity in Evolutionary Inference for Quality-Diversity Search

Hugging Face Daily Papers Papers
quality-diversity evolutionary-inference large-language-models heterogeneous-llms distributed-search core-war mutation-operators

Summary

DEI introduces a distributed Quality-Diversity search framework using heterogeneous LLMs as mutation operators, showing that model diversity improves performance over homogeneous parallel approaches. Evaluated on the Core War domain, a four-node heterogeneous ensemble achieves significant gains in QD-Score and coverage.

We present DEI: Diversity in Evolutionary Inference, a distributed Quality-Diversity (QD) search framework that assigns heterogeneous large language models (LLMs) as mutation operators across peer nodes communicating with non-blocking collective operations. Unlike homogeneous parallel search, which replicates a single model's inductive biases across all workers, DEI treats each LLM's distinct creative prior as a complementary source of behavioral novelty. Extending the Digital Red Queen framework with DEI, nodes share local optimal solutions at the end of each round to seed the next round's population. This creates cross-model adversarial pressure that drives robustness beyond intra-model self-play. Evaluated on the Core War domain, a competitive programming benchmark in which Redcode warrior programs battle inside a simulated machine, a four-node heterogeneous ensemble (GPT-5.4-mini, Claude Sonnet 4.6, GPT-5.2, and Claude Haiku 4.5) achieves 124 percent higher merged-archive QD-Score (45.90 vs. 20.46) and 28 percent higher coverage (80.6 percent vs. 63.0 percent of cells) than a single-node baseline at equal total LLM-call budget. The heterogeneous ensemble also outperforms an equally-budgeted homogeneous ensemble on QD-Score, coverage, and held-out solution generality across all four model families. These results provide the first empirical evidence that model diversity, not merely parallelism, is the key driver of gain in distributed LLM-based QD search.
Original Article
View Cached Full Text

Cached at: 06/10/26, 12:12 AM

Paper page - DEI: Diversity in Evolutionary Inference for Quality-Diversity Search

Source: https://huggingface.co/papers/2605.27130

Abstract

A distributed Quality-Diversity search framework uses heterogeneous large language models as mutation operators to enhance evolutionary inference, demonstrating that model diversity improves performance over homogeneous parallel approaches.

We present DEI: Diversity inEvolutionary Inference, a distributedQuality-Diversity(QD) search framework that assignsheterogeneous large language models(LLMs) asmutation operatorsacrosspeer nodescommunicating withnon-blocking collective operations. Unlike homogeneous parallel search, which replicates a single model’s inductive biases across all workers, DEI treats each LLM’s distinct creative prior as a complementary source of behavioral novelty. Extending theDigital Red Queen frameworkwith DEI, nodes sharelocal optimal solutionsat the end of each round to seed the next round’s population. This createscross-model adversarial pressurethat drives robustness beyond intra-model self-play. Evaluated on theCore Wardomain, a competitive programming benchmark in whichRedcode warrior programsbattle inside a simulated machine, a four-node heterogeneous ensemble (GPT-5.4-mini, Claude Sonnet 4.6, GPT-5.2, and Claude Haiku 4.5) achieves 124 percent higher merged-archiveQD-Score(45.90 vs. 20.46) and 28 percent highercoverage(80.6 percent vs. 63.0 percent of cells) than a single-node baseline at equal total LLM-call budget. The heterogeneous ensemble also outperforms an equally-budgeted homogeneous ensemble onQD-Score,coverage, andheld-out solution generalityacross all four model families. These results provide the first empirical evidence that model diversity, not merely parallelism, is the key driver of gain in distributed LLM-based QD search.

View arXiv pageView PDFGitHub2Add to collection

Get this paper in your agent:

hf papers read 2605\.27130

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.27130 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.27130 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.27130 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Where You Inject Diversity Matters: A Unified Framework for Diverse Generation

arXiv cs.CL

This paper introduces a unified framework for test-time diverse generation in large language models, categorizing methods by where diversity is injected (surface-level vs. specification-level). It proposes specification-level methods that generate diverse intermediate specifications, achieving better output diversity across five open-ended tasks and four backbone models while maintaining quality.

Reasoning or Memorization? Direction-Aware Diversity Exploration in LLM Reinforcement Learning

arXiv cs.AI

This paper introduces DiRL, a direction-aware reinforcement learning framework that distinguishes reasoning-driven diversity from memorization-driven diversity in LLM exploration. It extracts an internal reasoning-memorization direction from model representations and shapes rewards to prioritize reasoning-aligned exploration, showing improvements on math and general reasoning benchmarks.