RLHF Evolves Into Constitutional AI Training
The AI alignment landscape shifts as Constitutional AI methods begin replacing traditional RLHF, promising scalable and …
🏷️ rlhf
8 articles about 'rlhf'
Critical Jailbreak Flaw Hits All Major LLM Providers
Security researchers uncover a universal jailbreak vulnerability that bypasses safety guardrails across GPT-4, Claude, G…
OpenAI Proposes Recursive Reward Modeling
OpenAI researchers introduce Recursive Reward Modeling, a new alignment technique designed to keep advanced AI systems s…
OpenAI Proposes Constitutional AI Alternative
OpenAI researchers introduce a new alignment framework challenging Anthropic's Constitutional AI approach with rule-base…
Berkeley AI Lab Releases Open Source RLHF Framework
UC Berkeley's AI research lab publishes a comprehensive open source framework for RLHF training of large language models…
MIT Proposes New Constitutional AI Alignment Method
MIT researchers introduce a novel alignment framework that builds on Anthropic's Constitutional AI to improve safety in …
OpenAI Explores Constitutional AI Safety Methods
OpenAI has published new research on constitutional AI training, a safety approach pioneered by rival Anthropic, signali…
Constitutional AI Cuts Harmful Outputs by Up to 75%
New research shows Constitutional AI training methods dramatically reduce toxic and harmful outputs from large language …
1