DeepSurvey: Agent-Oriented Automated Survey Generation with Analytical Depth and Citation Reliability

arXiv:2605.29522 · cs.AI · Submitted 2026-05-28 · Read on arXiv

cs.AI

Submitted: 2026-05-28

Updated: 2026-09-29

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Agentic System — 1× in this paper · explained in 2 episodes
• Alignment — 1× in this paper
• Clustering — 1× in this paper · explained in 2 episodes
• Grounding — 1× in this paper · explained in 3 episodes
• Papers — 1× in this paper · explained in 3 episodes

Sources

• SurGE: A Benchmark and Evaluation Framework for Scientific Survey Generation
• PINNacle: A Comprehensive Benchmark of Physics-Informed Neural Networks for Solving PDEs
• LexRank: Graph-based Lexical Centrality as Salience in Text Summarization
• MinerU: An Open-Source Solution for Precise Document Content Extraction
• LLM$\times$MapReduce-V2: Entropy-Driven Convolutional Test-Time Scaling for Generating Long-Form Articles from Extremely Long Resources
• OpenAI GPT-5 System Card
• AutoSurvey: Large Language Models Can Automatically Write Surveys
• MVSS: A Unified Framework for Multi-View Structured Survey Generation
• SurveyG: A Multi-Agent LLM Framework with Hierarchical Citation Graph for Automated Survey Generation
• Deep Literature Survey Automation with an Iterative Workflow
• LiRA: A Multi-Agent Framework for Reliable and Readable Literature Review Generation
• Agentic AutoSurvey: Let LLMs Survey LLMs
• RecurrentGPT: Interactive Generation of (Arbitrarily) Long Text
• TaxoAlign: Scholarly Taxonomy Generation Using Language Models
• Agentic AI-Empowered Dynamic Survey Framework
• SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing
• Hierarchical Catalogue Generation for Literature Review: A Benchmark
• AutoSurvey2: Empowering Researchers with Next Level Automated Literature Surveys
• SurveyX: Academic Survey Automation via Large Language Models
• OUTLINEFORGE: Hierarchical Reinforcement Learning with Explicit States for Scientific Writing

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems