SurveyX: Academic Survey Automation via Large Language Models

arXiv 2025 55.2 method

TLDR

SurveyX automates academic survey generation using LLMs with two-phase decomposition, online retrieval, and AttributeTree, outperforming existing systems.

Reasoning

The paper presents a clear methodology and evaluation, showing improvements in content and citation quality. However, it focuses narrowly on survey generation rather than broader scientific discovery, and lacks details on limitations or generalizability.

Read-first score

Read-first score 55.2, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 59.

Recency 8%

86.7

Uses a gentle age decay so recent papers surface without erasing older foundations. 2025

Methodology quality 25%

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

Topical relevance 42%

49.2

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

Reproducibility 25%

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

Field roles

FrontierMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 19.

Keyword Scores

survey generation

literature review agent

paper writing agent

automated research

research automation

AI for scientific research

automated scientific discovery

scientific discovery agent

AI scientist

autonomous research agent

automated experimentation

experiment design agent

Deep Analysis

Innovations

Decomposition of survey generation into Preparation and Generation phases
Online reference retrieval
AttributeTree pre-processing method
Re-polishing process

Methodology

SurveyX decomposes automated survey generation into Preparation and Generation phases, incorporating online reference retrieval, an AttributeTree pre-processing method, and a re-polishing process to enhance content and citation quality. The system is evaluated against existing automated survey generation systems using content quality and citation quality metrics.

Key Results

SurveyX outperforms existing systems with a 0.259 improvement in content quality and a 1.76 enhancement in citation quality, approaching human expert performance across multiple evaluation dimensions.

Resources

Paper