This page is still being polished. If you have thoughts, please share them via the feedback form.
Data on this page is preliminary and may change. Please do not share or cite these figures publicly.
Input validation, output filtering, and content moderation classifiers.
Also in Non-Model
Many researchers employ in-context learning methods either to induce LLMs to generate unethical content and subsequently correct them (Duan et al., 2024b) or to guide LLMs in recognizing unethical instructions within the prompts (Roy et al., 2023; Phute et al., 2024; Ganguli et al., 2023), thereby preventing the generation of unethical outputs.
Zhang et al. (2024a) introduce Intention Analysis (IA), a robust defense strategy implemented through a two-stage approach. Firstly, the model identifies the user’s underlying intentions by answering a designed question, followed by generating responses aligned with the recognized intentions. It is observed that the model’s defensive capabilities significantly improve upon identifying negative intentions. Another approach employs a multi-step reasoning method, where Theory-guided Instructions are incorporated into prompts to guide LLMs in producing ethically aligned content (Phute et al., 2024). DeNEVIL (Duan et al., 2024b) is designed to dynamically exploit value vulnerabilities in LLMs and use corrected generations to train LLMs for ethical alignment. In addition to leveraging the intrinsic knowledge of LLMs to guide value principles, OPO (Xu et al., 2023a) employs external memory of moral rule knowledge to constrain the outputs generated from LLMs.
Reasoning
Output filtering mechanism blocks amoral model responses without modifying training or weights.
Value Misalignment
99.9 OtherValue Misalignment > Mitigating social bias
1 AI SystemValue Misalignment > Privacy protection
1 AI SystemValue Misalignment > Methods for mitigating toxicity
1 AI SystemValue Misalignment > Methods for mitigating LLM amorality
1 AI SystemRobustness to attack
1 AI SystemLarge Language Model Safety: A Holistic Survey
Shi, Dan; Shen, Tianhao; Huang, Yufei; Li, Zhigen; Leng, Yongqi; Jin, Renren; Liu, Chuang; Wu, Xinwei; Guo, Zishan; Yu, Linhao; Shi, Ling; Jiang, Bojian; Xiong, Deyi (2024)
The rapid development and deployment of large language models (LLMs) have introduced a new frontier in artificial intelligence, marked by unprecedented capabilities in natural language understanding and generation. However, the increasing integration of these models into critical applications raises substantial safety concerns, necessitating a thorough examination of their potential risks and associated mitigation strategies. This survey provides a comprehensive overview of the current landscape of LLM safety, covering four major categories: value misalignment, robustness to adversarial attacks, misuse, and autonomous AI risks. In addition to the comprehensive review of the mitigation methodologies and evaluation resources on these four aspects, we further explore four topics related to LLM safety: the safety implications of LLM agents, the role of interpretability in enhancing LLM safety, the technology roadmaps proposed and abided by a list of AI companies and institutes for LLM safety, and AI governance aimed at LLM safety with discussions on international cooperation, policy proposals, and prospective regulatory directions. Our findings underscore the necessity for a proactive, multifaceted approach to LLM safety, emphasizing the integration of technical solutions, ethical considerations, and robust governance frameworks. This survey is intended to serve as a foundational resource for academy researchers, industry practitioners, and policymakers, offering insights into the challenges and opportunities associated with the safe integration of LLMs into society. Ultimately, it seeks to contribute to the safe and beneficial development of LLMs, aligning with the overarching goal of harnessing AI for societal advancement and well-being. A curated list of related papers has been publicly available at https://github.com/tjunlp-lab/Awesome-LLM-Safety-Papers.
Build and Use Model
Training, fine-tuning, and integrating the AI model
Deployer
Entity that integrates and deploys the AI system for end users
Manage
Prioritising, responding to, and mitigating AI risks
