EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery
EvoDuet co-evolves solutions and web queries via a retrieval gate to boost LLM discovery gains up to 82.3% across optimization tasks.
Published Sep 30, 2026▲ 109 on Hugging FaceCode ★ 4arXiv ↗
Only vote on papers you've read. Sign in with GitHub to vote.
EvoDuet delivers impressive gains via co-evolved queries and a retrieval gate, yet its fixed-parameter framework risks overfitting, hides evaluation costs, and collapses on smaller models like Qwen, leaving its true mechanism and scalability unresolved.
Abstract
Evolutionary search with large language models (LLMs) can stall when progress requires external knowledge the model lacks. Supplying relevant documents helps, but simply adding web search tool can keep returning the same pages as solutions change. We introduce EvoDuet, a bi-level optimization method that co-evolves solutions and search queries with fixed model parameters. At each iteration, a retrieval gate lets the LLM assess its knowledge gap and choose to retrieve new documents, reuse stored ones, or proceed without them. An inner loop refines queries and ranks documents by the solution scores they are predicted to yield; an outer loop generates candidates in parallel from these documents and records the evaluated outcomes for later searches. Across 21 optimization tasks with one candidate per iteration, EvoDuet raises OpenEvolve's normalized discovery gain from 74.1% to 78.0% with GPT-5.6-Luna and from 61.3% to 82.3% with Gemini-3.8-Flash, whereas Qwen3.5-9B does not benefit. Our best runs surpass the previously reported best scores on eight tasks, including Swap Reduction on Q20 and Rosetta, and match them on three more. EvoDuet also improves with other scaffolds (e.g., Top-K, EvoX) on Sums/Diffs and Denoising, demonstrating its applicability across evolutionary search scaffolds.