My wife and I were at a married couples counseling meeting. The speaker was asking the audience random questions. Then he called on me and said, "Sir do you know what your wife's favorite LLM is?" Feeling kind of singled out, I turned to my wife and said, "It's Llama, isn't
LLMS
-
MoE Inference Benefits: Why Mixture of Experts Improves Performance
By
–
It can be unintuitive why the Transformer-style MoE (in Mixtral/GPT4) has inference benefits.
Dima simplifies it with a clear explanation showcasing that MoE help inference once there's sufficient volume of requests (which hopefully are diverse enough that they don't hit the same -
Advanced Open Source Models and the Turing Test Challenge
By
–
No question about it. It will be interesting to see if some of the advanced open source models that have been released recently could also be fine-tuned to pass it. And I have no doubt that Turing Test has long ago been passed internally.
-
Challenges of long-term memory and intent management in AGI
By
–
Wondering how would it look like in case a long term memory will become a thing But the complexity will just keep growing, even in case of AGI, you would need to learn how to operate its intents in a right way
-
Scale and NVIDIA Collaborate on NeMo SteerLM for LLMs
By
–
Scale collaborates with NVIDIA to power the next generation of LLMs with NeMo SteerLM https://
bit.ly/47Hy1f6 #AI #MachineLearning #DeepLearning #LLMs #DataScience -
LLMs Understanding Time and Place with OS Integration
By
–
Can LLMs understand time and place?
— Aaron Ng (@localghost) 10 décembre 2023
Here’s @apolloaiapp reading my location and calendar availability to propose a calendar event. A hint of what LLMs look like once they start working with your OS. pic.twitter.com/n4K3rTzXvCCan LLMs understand time and place? Here’s @apolloaiapp reading my location and calendar availability to propose a calendar event. A hint of what LLMs look like once they start working with your OS.
-
AnythingLLM: Private Open-Source Enterprise Document Chatbot Solution
By
–
💬AnythingLLM: a private ChatGPT to chat with anything
— LangChain (@LangChain) 10 décembre 2023
An efficient🚤, customizable🧩, and open-source enterprise-ready document chatbot solution.
An awesome example of a fullstack (JS) LLM application (s/o @MintplexLabs)
Code: https://t.co/GVzrH1N6hR pic.twitter.com/dCFiIRErFYAnythingLLM: a private ChatGPT to chat with anything An efficient, customizable, and open-source enterprise-ready document chatbot solution. An awesome example of a fullstack (JS) LLM application (s/o @MintplexLabs
) Code: https://
github.com/Mintplex-Labs/
anything-llm
… -
Embeddings: Essential Foundations for AI and Machine Learning
By
–
Embeddings: What they are and why they matter https://
bit.ly/3RcPTaJ #AI #MachineLearning #DeepLearning #LLMs #DataScience -
LLM Efficiency Spectrum: Algorithmic Advancements Review
By
–
10/ The Efficiency Spectrum of LLMs – a comprehensive review of algorithmic advancements aimed at improving LLM efficiency.
-
Outstanding Paper Award for Machine Translation Evaluation Metrics
By
–
Congratulations to @_danieldeutsch
, George Foster, and @markuseful
, co-authors of the paper “Ties Matter: Meta-Evaluating Modern Metrics with Pairwise Accuracy and Tie Calibration”, for winning the #EMNLP2023 Outstanding Paper Award in Machine Translation!