The latest #AI news are scary
→ An AI reads thoughts…
→ A #ChatGPT for armies…
→ @geoffreyhinton leaves Google because AI goes too far…
We decipher together the big news #ArtificialIntelligence: https://youtu.be/I8NiQlAry4E
LLMS
-
Scary AI News: thoughts, army, Hinton leaves Google
By
–
-
ResiDual Solves Gradient Vanishing and Representation Collapse
By
–
Optimizing Transformers: Microsoft & RUC’s ResiDual Solves Gradient Vanishing and Representation Collapse Issues
-
Flan-T5 Performance Analysis: Efficiency Across Language Models
By
–
Pretty cool idea! Great to see Flan-T5 (despite being the smallest model here) hold it's ground pretty well . It even outperforms other LMs like Dolly or StableLM. Also another noteworthy point is that at "compute-match", Flan-T5 3B is equivalent to the cost of a 1.5B
-
Evolution of Human-Computer Interfaces: From Punch Cards to Natural Language
By
–
oh that wasn’t very clear—i meant the evolution of how we use computers,from punch cards to command lines to higher-level programming to GUIs to touch interfaces to programming in natural language
-
Chain-of-Thought Prompting: Emergence in Large Language Models
By
–
Additional response 2: Another popular example of emergence which also underscores qualitative changes in the model is chain-of-thought prompting, for which performance is worse than answering directly for small models, but much better than answering directly for large models.
-
U-shaped Performance: Emergent Abilities in Model Scaling
By
–
Additional response 1: Many emergent abilities also cannot be explained by these arguments. Consider the below plots which show a U-shaped phenomena: performance actually decreases for several model scales, until it suddenly spikes up again.
-
Emergent Abilities Still Visible When Plotted
By
–
Response 2A: This is untrue, at least for many emergent abilities. You can plot it and still see the emergent spike.
-
How To Access ChatGPT-4 For Free Without Plus Subscription
By
–
How To Use #ChatGPT-4 For Free (Without A Plus Subscription): https://
slashgear.com/1272430/how-to
-use-chatgpt-4-for-free-without-plus-subscription/
…
————
#GenerativeAI #AI #GPT4 #GPT3 #NLU #NLG #NLProc #ConversationalAI #MachineLearning #DeepLearning #DataScience #DataScientists -
Data Stack Optimization Race for AI Models and Applications
By
–
So once again we have another divisive topic here as there's a new race to recreate the data stack in a way that's optimized for AI as large language and diffusion models enter every aspect of our lives—from enterprise tech to consumer applications.
-
MIT Professor Breaks Down Large Language Models Part Two
By
–
MIT professor breaks down large language models, part two: http://
rb.gy/4givx