With Claude 3 APIs available, I would love to see more academics re-running their tests (like the papers on AIs performance in medicine & law). Advantages:
1) Helps us understand the variance among GPT-4 class models
2) Helps us understand if there are universal LLM shortcomings
MACHINE LEARNING
-
Academics Should Re-test LLM Performance Across Domains
By
–
-
Octopus v2: On-Device AI Agents Surpass GPT-4 Accuracy
By
–
Stanford researchers just introduced Octopus v2, a groundbreaking new framework for on-device AI agents. It outperforms GPT-4 in accuracy by fine-tuning language models with special functional tokens. The new era of on-device AI agents is coming. https://
x.com/nexa4ai/status
/1775993692509323582/video/1
… -
AI Chatbots Transform Sustainability Profession
By
–
How #AI #Chatbots are transforming the #Sustainability profession
by @makower @GreenBiz Read more: https://
buff.ly/43utXgi #BigData #ArtificialIntelligence #MachineLearning #DataScience #Tech cc: @bigdata @patrickgunz_ch @ravikikan -
GPT-4 vs ChatGPT: AI Chatbot Comparison Analysis
By
–
GPT-4 vs. ChatGPT: #AI Chatbot Comparison
by Drew Robb @eWEEKNews Read more: https://
buff.ly/3zXvDl9 #MachineLearning #ArtificialIntelligence #ML #MI #DataScience #Chatbots cc: @pascal_bornet @pbalakrishnarao @rtehrani -
AI Leaders Network at Exclusive Dinner Event
By
–
Thanks @osanseviero and @huggingface for inviting me to a wonderful AI dinner, where I had the pleasure of catching up with old friends, @sarahookr @neilzegh @laurentsifre @ylecun
, meet new amazing people, and do one of the things I absolutely love: brainstorm about AI with -
Exploring the Latent Space of Agent Architectures
By
–
more options are good to explore the latent space of agent archs 🙂
-
AI Targeting System Lavender: Machine Learning Military Application
By
–
Yes, the book written by the commander of the unit that runs Lavender (etc.), a book that "offers a short guide to building a “target machine,” similar in description to Lavender, based on AI and machine-learning algorithms." Whose description then matches what he discusses in
-
Synthetic Data Pre-caching Issues in ML Architecture
By
–
i think it is a data issue, since synthetic data induces pre-caching with the same arch
-
Activation Trade-offs: Current vs Future Token Optimization
By
–
I guess there is a tradeoff between how each activation is used, whether it is doing something that's optimal for a current or future token. i'm using information and store both pretty loosely. 🙂
-
Transformers Token Processing: Information Storage Nuances
By
–
yes! there is a lot of nuance. My preferred way to put it is "transformers don't pre-store information for future tokens at the expense of the current token" (very much)