Awesome Auto Research Hub Papers · Datasets · Projects
← Back to papers

NanoResearch: Co-Evolving Skills, Memory, and Policy for Personalized Research Automation

arXiv 2026 52 method, system

TLDR

NanoResearch is a multi-agent framework for personalized research automation using tri-level co-evolution of skills, memory, and policy.

Reasoning

The paper addresses a critical gap in research automation by focusing on personalization, proposing a novel tri-level co-evolution mechanism. However, the abstract is cut off, limiting visibility into experimental details and real-world validation.

Read-first score

Read-first score 52, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 68.

Recency 8%
100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Topical relevance 42%
56.7

Uses existing LLM keyword relevance scores normalized to 0-100. AI scientist,automated scientific discovery,autonomous research agent,automated research,literature review agent,survey generation,automated experimentation,experiment design agent,AI for scientific research,paper writing agent,research automation,scientific discovery agent

Methodology quality 25%
50

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=experiment

Reproducibility 25%
30

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 99.

Keyword Scores

automated research
9
research automation
9
AI for scientific research
8
paper writing agent
8
automated scientific discovery
7
autonomous research agent
7
AI scientist
6
scientific discovery agent
6
automated experimentation
3
literature review agent
2
experiment design agent
2
survey generation
1

Deep Analysis

Innovations

  • Skill bank that distills recurring operations into compact procedural rules reusable across projects
  • Memory module that maintains user- and project-specific experience across sessions
  • Label-free policy learning that converts free-form feedback into persistent parameter updates of the planner
  • Tri-level co-evolution of skills, memory, and policy for personalized research automation

Methodology

NanoResearch is a multi-agent framework with a skill bank, memory module, and label-free policy learning that co-evolve. The skill bank distills recurring operations into reusable rules, the memory module retains user/project experience, and policy learning updates planner parameters from free-form feedback. Experiments compare against state-of-the-art AI research systems.

Key Results

NanoResearch achieves substantial gains over state-of-the-art AI research systems and progressively refines itself to produce better research at lower cost over successive cycles.

Tags

AI