Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Agentic AI-Empowered Dynamic Survey Framework

Furkan Mumcu, Lokman Bekit, Michael J. Jones, Anoop Cherian, Yasin Yilmaz

TL;DR

This work reframes survey writing as a long-horizon maintenance problem and introduces the agentic Dynamic Survey Framework, which treats surveys as living documents with a persistent state S_t=(D_t,C) and a frozen structure C to prevent semantic drift. Updates from new literature are integrated through a conservative, hierarchical multi-agent loop (Outline, Analysis, Routing, Synthesis) that emphasizes locality and abstention to minimize disruption, formalized by a disruption objective L(S_t,S_{t+1}). Retrospective experiments across five diverse surveys demonstrate high routing accuracy, strong semantic alignment, and zero out-of-scope edits, with modest token changes per update, indicating effective integration of new work without destabilizing existing content. Across backbone models, the framework maintains robust performance, though results also reveal that architectural constraints and agent specialization matter more than model size. The study argues that dynamic survey maintenance can complement traditional scholarly practice, offering practical guidance on deployment, human oversight, and versioning to manage living surveys over time.

Abstract

Survey papers play a central role in synthesizing and organizing scientific knowledge, yet they are increasingly strained by the rapid growth of research output. As new work continues to appear after publication, surveys quickly become outdated, contributing to redundancy and fragmentation in the literature. We reframe survey writing as a long-horizon maintenance problem rather than a one-time generation task, treating surveys as living documents that evolve alongside the research they describe. We propose an agentic Dynamic Survey Framework that supports the continuous updating of existing survey papers by incrementally integrating new work while preserving survey structure and minimizing unnecessary disruption. Using a retrospective experimental setup, we demonstrate that the proposed framework effectively identifies and incorporates emerging research while preserving the coherence and structure of existing surveys.

Agentic AI-Empowered Dynamic Survey Framework

TL;DR

This work reframes survey writing as a long-horizon maintenance problem and introduces the agentic Dynamic Survey Framework, which treats surveys as living documents with a persistent state S_t=(D_t,C) and a frozen structure C to prevent semantic drift. Updates from new literature are integrated through a conservative, hierarchical multi-agent loop (Outline, Analysis, Routing, Synthesis) that emphasizes locality and abstention to minimize disruption, formalized by a disruption objective L(S_t,S_{t+1}). Retrospective experiments across five diverse surveys demonstrate high routing accuracy, strong semantic alignment, and zero out-of-scope edits, with modest token changes per update, indicating effective integration of new work without destabilizing existing content. Across backbone models, the framework maintains robust performance, though results also reveal that architectural constraints and agent specialization matter more than model size. The study argues that dynamic survey maintenance can complement traditional scholarly practice, offering practical guidance on deployment, human oversight, and versioning to manage living surveys over time.

Abstract

Survey papers play a central role in synthesizing and organizing scientific knowledge, yet they are increasingly strained by the rapid growth of research output. As new work continues to appear after publication, surveys quickly become outdated, contributing to redundancy and fragmentation in the literature. We reframe survey writing as a long-horizon maintenance problem rather than a one-time generation task, treating surveys as living documents that evolve alongside the research they describe. We propose an agentic Dynamic Survey Framework that supports the continuous updating of existing survey papers by incrementally integrating new work while preserving survey structure and minimizing unnecessary disruption. Using a retrospective experimental setup, we demonstrate that the proposed framework effectively identifies and incorporates emerging research while preserving the coherence and structure of existing surveys.
Paper Structure (83 sections, 16 equations, 3 figures, 13 tables, 1 algorithm)

This paper contains 83 sections, 16 equations, 3 figures, 13 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Dynamic Survey Framework
  4. Surveys as Living Documents
  5. Agentic Components Overview
  6. Dynamic Survey Update Loop
  7. Experiments
  8. Retrospective Survey Maintenance Benchmark
  9. Evaluation Metrics
  10. Baselines and Backbone Models
  11. Main Results
  12. Human Evaluation
  13. Robustness across Backbone Models
  14. Ablation Study
  15. Conclusion
  16. ...and 68 more sections

Figures (3)

  • Figure 1: Static survey writing versus dynamic survey maintenance. Traditional surveys capture a fixed snapshot of the literature and degrade as new papers are published, whereas dynamic surveys are continuously updated to integrate new work while maintaining a stable structure and narrative.
  • Figure 2: Overview of the Dynamic Survey Framework. Incoming papers are analyzed, filtered, and selectively routed for localized synthesis into existing survey sections or tables.
  • Figure 2.a: BLEU