Tesla Recalls Autopilot Software in 2 Million Vehicles:
SAFETY
-
RLHF and RLAIF: Training Techniques Revolutionizing Language Models
By
–
Discover the magic behind ChatGPT's effectiveness in our deep dive into RLHF and its innovative counterpart, RLAIF. Learn how these training techniques are revolutionizing language models, making them safer, smarter, and more efficient.
-
New RLHF Short Course for Large Language Model Alignment
By
–
New short course on Reinforcement Learning from Human Feedback! RLHF is one of the key techniques that led to the rise of modern LLMs. It is used to align LLMs with human preferences, to make them more honest, helpful and harmless, by
— Andrew Ng (@AndrewYNg) 13 décembre 2023
(i) learning a reward function that mimics… pic.twitter.com/hmA5P3owV8New short course on Reinforcement Learning from Human Feedback! RLHF is one of the key techniques that led to the rise of modern LLMs. It is used to align LLMs with human preferences, to make them more honest, helpful and harmless, by (i) learning a reward function that mimics
-
Google Introduces DICES Dataset for Conversational AI Safety
By
–
Stop by the #NeurIPS2023 Google booth today at 9:15am where @laroyo
, @vinodkpg
, & @AliciaVParrish will introduce the DICES dataset, a shared resource and benchmark that respects diverse perspectives during safety evaluation of conversational AI systems. -
Generative AI Safety Best Practices at NeurIPS
By
–
Working on generative AI models? At 9:15am today, Diana Mincu & Kevin Robinson will be at the @NeurIPSConf Google booth to talk about best practices for generative AI safety.
-
Evaluating Auto-Regressive LLMs: Testing Bias and Prompt Injection
By
–
Ways to get Auto-Regressive LLMs to produce correct answers almost all the time:
– test on the training set.
– write the answer in the prompt. -
Google Gemini obscures AI technology details from public scrutiny
By
–
Google’s Gemini continues the dangerous obfuscation of AI technology Google’s Gemini project discloses almost no relevant technical details, hiding from society what’s happening in the cloud. https://
zdnet.com/article/google
s-gemini-continues-the-dangerous-obfuscation-of-ai-technology/
… @GoogleDeepMind @GoogleAI -
Seeing Reality Through AI: Beyond Imagination and Bias
By
–
The ability to see through the lens of reality and not what you've conjured up in your imagination.
-
WSJ Reviews The Coming Wave: AI Wonders and Troubles Ahead
By
–
review of http://
the-coming-wave.com from @WSJ https://
wsj.com/arts-culture/b
ooks/the-coming-wave-review-wonders-ahead-trouble-too-27e2fb18?mod=arts-culture_lead_story
…