LLM06
Sensitive Information Disclosure
LLMs leaking training data, PII, prompts
233 papers Browse all papers
Monthly publications
Paper types
defense 91
attack 69
benchmark 56
survey 12
tool 5
Domains
nlp 229
multimodal 19
vision 13
federated-learning 7
generative 5
graph 4
audio 2
tabular 1
Co-occurring categories
Other OWASP categories that appear on the same papers
ML03 Model Inversion Attack
83 LLM01 Prompt Injection
70 ML04 Membership Inference Attack
34 LLM07 Insecure Plugin Design
8 LLM08 Excessive Agency
8 ML02 Data Poisoning Attack
7 ML05 Model Theft
7 ML10 Model Poisoning
5 LLM03 Training Data Poisoning
3 ML06 AI Supply Chain Attacks
3 ML01 Input Manipulation Attack
3 ML09 Output Integrity Attack
2 LS06 Red-Team Agents
2 LS03 Reconnaissance & OSINT
1 ML07 Transfer Learning Attack
1Top cited papers
1152837465565758494104
Language Models are Injective and Hence Invertible
2025 attack
Eliciting Secret Knowledge from Language Models
2025 benchmark
Mitigating the OWASP Top 10 For Large Language Models Applications using Intelligent Agents
2026 defense
Hubble: a Model Suite to Advance the Study of LLM Memorization
2025 benchmark
Extracting books from production language models
2026 attack
ConVerse: Benchmarking Contextual Safety in Agent-to-Agent Conversations
2025 benchmark
You Can't Steal Nothing: Mitigating Prompt Leakages in LLMs via System Vectors
2025 defense
SALT: Steering Activations towards Leakage-free Thinking in Chain of Thought
2025 defense
Confidential LLM Inference: Performance and Cost Across CPU and GPU TEEs
2025 benchmark
Extracting alignment data in open models
2025 attack