ResearchEVO: An End-to-End Framework for Automated Scientific Discovery and Documentation

2026-04-07

AI Summary

This work proposes the first end-to-end automated scientific research framework capable of autonomously discovering novel scientific phenomena from unguided experiments and generating interpretable theories alongside complete academic papers. By integrating a large language model (LLM)-guided dual-dimensional coevolutionary search, sentence-level retrieval-augmented generation, hallucination mitigation through verification, and automated experimental design, the system achieves full automation from algorithmic discovery to literature-grounded scholarly writing. Evaluated on quantum error correction and physics-informed neural network tasks, the framework successfully uncovers new, interpretable algorithms and produces fully compilable LaTeX manuscripts with zero fabricated citations, thereby overcoming the limitation of conventional AI-assisted research tools that operate only within isolated stages of the scientific workflow.

Abstract

An important recurring pattern in scientific breakthroughs is a two-stage process: an initial phase of undirected experimentation that yields an unexpected finding, followed by a retrospective phase that explains why the finding works and situates it within existing theory. We present ResearchEVO, an end-to-end framework that computationally instantiates this discover-then-explain paradigm. The Evolution Phase employs LLM-guided bi-dimensional co-evolution -- simultaneously optimizing both algorithmic logic and overall architecture -- to search the space of code implementations purely by fitness, without requiring any understanding of the solutions it produces. The Writing Phase then takes the best-performing algorithm and autonomously generates a complete, publication-ready research paper through sentence-level retrieval-augmented generation with explicit anti-hallucination verification and automated experiment design. To our knowledge, ResearchEVO is the first system to cover this full pipeline end to end: no prior work jointly performs principled algorithm evolution and literature-grounded scientific documentation. We validate the framework on two cross-disciplinary scientific problems -- Quantum Error Correction using real Google quantum hardware data, and Physics-Informed Neural Networks -- where the Evolution Phase discovered human-interpretable algorithmic mechanisms that had not been previously proposed in the respective domain literatures. In both cases, the Writing Phase autonomously produced compilable LaTeX manuscripts that correctly grounded these blind discoveries in existing theory via RAG, with zero fabricated citations.

Problem

• scientific discovery
• automated research
• algorithm evolution
• scientific documentation
• LLM-guided exploration

Innovation

• LLM-guided co-evolution
• automated scientific discovery
• retrieval-augmented generation
• anti-hallucination verification
• end-to-end research automation