Evaluating Novelty in AI-Generated Research Plans Using Multi-Workflow LLM Pipelines

TL;DR

This paper evaluates whether multi-step, agentic workflows using Large Language Models can produce more novel and feasible research plans compared to single-step prompting, highlighting the importance of workflow design in AI research ideation.

Contribution

It introduces and benchmarks multiple multi-workflow LLM architectures for generating research plans, demonstrating their superior novelty and feasibility over simpler methods.

Findings

Decomposition-based workflows achieve high novelty scores (4.17/5).

Reflection-based approaches score significantly lower (2.33/5).

High-performing workflows maintain feasibility across domains.

Abstract

The integration of Large Language Models (LLMs) into the scientific ecosystem raises fundamental questions about the creativity and originality of AI-generated research. Recent work has identified ``smart plagiarism'' as a concern in single-step prompting approaches, where models reproduce existing ideas with terminological shifts. This paper investigates whether agentic workflows -- multi-step systems employing iterative reasoning, evolutionary search, and recursive decomposition -- can generate more novel and feasible research plans. We benchmark five reasoning architectures: Reflection-based iterative refinement, Sakana AI v2 evolutionary algorithms, Google Co-Scientist multi-agent framework, GPT Deep Research (GPT-5.1) recursive decomposition, and Gemini~3 Pro multimodal long-context pipeline. Using evaluations from thirty proposals each on novelty, feasibility, and impact, we find that decomposition-based and long-context workflows achieve mean novelty of 4.17/5, while reflection-based approaches score significantly lower (2.33/5). Results reveal varied performance across research domains, with high-performing workflows maintaining feasibility without sacrificing creativity. These findings support the view that carefully designed multi-stage agentic workflows can advance AI-assisted research ideation.