Good morning fellow AI enthusiast! This is the seventh video of my LLM series for our free course "Training & Fine-Tuning LLMs for Production"! In this really exciting one, we dive into RLHF and its recent alternative RLAIF, where humans are once again replaced by more AI!
RESEARCH
-
Air Frying as Secret to Training Language Models
By
–
air frying is the secret to cooking language models right
-
Stanford’s DetectGPT Identifies AI-Generated Content with 95% Accuracy
By
–
ChatGPT rocked the world with its ability to answer questions and write essays. Stanford researchers responded with a key guardrail concept, DetectGPT, a tool that can identify authorship with 95% accuracy.
-
Layoffs at Big Tech Companies Threaten Cutting-Edge AI Research
By
–
This is really bad. Not only because of the layoffs, but because a lot of cutting edge research these days is done at the big COs. I expect this will primarily hurt the “blue-sky” research. Much more pressure to keep only research that directly contributes to the bottom line.
-
Training Massive Text Embedding Model with Many Parameters
By
–
also training a massive text embedding model (many tokens AND many parameters) !
-
Improving vec2text Models with More Parameters and Tokens
By
–
shameless plug, training better (more parameters or more tokens) vec2text models would also be nice
-
Tokenization-Free Transformers: Scaling Breakthrough Research Idea
By
–
machine learning research question:
what’s an idea that you think would catch on, if only someone spent the money to test it at scale? i’ll go first: tokenization-free transformers -
Chain Rule Connections: Probability, Calculus, and Thought
By
–
Chain rule of probability, chain rule of calculus, chain of thought.
