Survey details LLM's dual role in generating and mitigating harmful content

A new survey paper titled "Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM" systematically reviews the dual nature of Large Language Models (LLMs). The paper explores how LLMs can both generate harmful content, such as toxic or biased language, and also serve as tools for safety through detection and prevention. It proposes a taxonomy of LLM harms and defenses, analyzes adversarial attacks, and evaluates mitigation strategies like RLHF and prompt engineering. AI

IMPACT Provides a comprehensive overview of LLM safety challenges and mitigation techniques, guiding future research in ethical AI development.

RANK_REASON The cluster contains a survey paper on LLM safety and harmful content generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• Chi Zhang
• Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM
• Hugging Face
• Large Language Models
• reinforcement learning from human feedback

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →