The resources related to the trustworthiness of large models (LMs) across multiple dimensions (e.g., safety, security, and privacy), with a special focus on multi-modal LMs (e.g., vision-language models and diffusion models).

• This repo is in progress 🌱 (manually collected).

• Badges:

  • Model:

    • 
    • 
    • 
    • 

  • Comment: ...

  • Venue: ...

• 🌻 Welcome to recommend resources to us via pulling requests or opening issues with the following format:

• [2025.01.09] 🎂 Happy 1st Birthday to Awesome-LM-SSP! Keep Going! 💪
• [2024.01.09] 🚀 LM-SSP is released!

• Book (3)
• Competition (5)
• Leaderboard (5)
• Toolkit (13)
• Survey (39)
• Paper (2128)
  • A. Safety (1080)
    • A0. General (28)
    • A1. Jailbreak (482)
    • A2. Alignment (127)
    • A3. Deepfake (88)
    • A4. Ethics (5)
    • A5. Fairness (60)
    • A6. Hallucination (114)
    • A7. Prompt Injection (93)
    • A8. Toxicity (83)
  • B. Security (399)
    • B0. General (14)
    • B1. Adversarial Examples (102)
    • B2. Agent (97)
    • B3. Poison & Backdoor (163)
    • B4. System (23)
  • C. Privacy (649)
    • C0. General (50)
    • C1. Contamination (15)
    • C2. Data Reconstruction (57)
    • C3. Data Reconstruction (4)
    • C4. Membership Inference Attacks (57)
    • C5. Model Extraction (13)
    • C6. Privacy-Preserving Computation (126)
    • C7. Property Inference Attacks (7)
    • C8. Side-Channel (10)
    • C9. Unlearning (66)
    • C10. Watermark & Copyright (244)

• Organizers: Tianshuo Cong (丛天硕), Xinlei He (何新磊), Zhengyu Zhao (赵正宇), Yugeng Liu (刘禹更), Delong Ran (冉德龙)

• This project is inspired by LLM Security, Awesome LLM Security, LLM Security & Privacy, UR2-LLMs, PLMpapers, EvaluationPapers4ChatGPT