可视化分析 Analytics

基于 69 篇论文的统计 69 papers analyzed

← 返回报告 ← Back to report
→ 研究空白发现（填表法） → Research Gap Discovery (Fill in the Blank) → 基准测试全景 → Benchmark Landscape

年份分布 Year Distribution

2005
1
2013
1
2019
1
2020
1
2021
2
2023
2
2024
18
2025
37
2026
6

论文类型 Paper Type

system
46
benchmark
8
empirical-study
7
survey
6
defense
1
position-paper
1

研究范围 Research Scope

penetration-testing
40
ctf
13
general-cybersecurity
5
vulnerability-assessment
5
defense
4
red-teaming
2

Agent 架构 Agent Framework

single-agent
32
multi-agent
21
none
10
hierarchical
3
human-in-the-loop
3

规划策略 Planning Strategy

ReAct
25
none
17
hierarchical-planning
5
RL-based
5
classical-planning
4
chain-of-thought
4
FSM
3
prompt-chaining
2
game-theoretic
2
tree-of-thought
1
POMDP-based
1

自动化程度 Automation Level

fully-autonomous
55
human-in-the-loop
6
semi-autonomous
6
copilot
2

攻击阶段覆盖 Attack Phase Coverage

每篇论文覆盖的渗透测试阶段 Papers covering each penetration testing phase

59 reconnaissance
51 scanning
58 enumeration
64 exploitation
26 post exploitation
32 privilege escalation
17 lateral movement
14 reporting