They are also frequently the best performing LLMs it seems on hugging face https://
reddit.com/r/LocalLLaMA/c
omments/17fhp9k/huge_llm_comparisontest_39_models_tested_7b70b/
…
RESEARCH
-
Best Performing LLMs Compared on Hugging Face
By
–
-
Real-time Rendering with Gaussian-SLAM and Efficient Splatting
By
–
Real-time Rendering
— AlphaSignal AI (@AlphaSignalAI) 17 décembre 2023
Scenes reconstructed with Gaussian-SLAM can be rendered in real-time, thanks to efficient Gaussian Splatting rendering. pic.twitter.com/K7FisSQk5DReal-time Rendering Scenes reconstructed with Gaussian-SLAM can be rendered in real-time, thanks to efficient Gaussian Splatting rendering.
-
Brown Researchers Develop Natural Language Robot Control System
By
–
researchers from Brown work on breaking human-robot barrier: new system translates natural language into robot actions
-
US Propaganda Behind AI Deceleration Narrative Debate
By
–
Imho the whole decel talk is just US propaganda
-
Student Use Cases for AI in Machine Learning and Data Science
By
–
Student Use Cases for AI https://
bit.ly/486OsRQ #AI #MachineLearning #DeepLearning #LLMs #DataScience -
Something Will Happen: Probability Equals One
By
–
This might be an unpopular opinion, but p(something happens) = 1
-
What Great Data Analysts Do and Why Organizations Need Them
By
–
What Great Data Analysts Do — and Why Every Organization Needs Them. #BigData #Analytics #DataScience #AI #ML #IoT #PyTorch #Python #RStats #TensorFlow #Java #JavaScript #ReactJS #Serverless #DataScientist #Linux #Programming #tech https://
bit.ly/3RKySGf -
Engineering Features for Building High Quality AI Models
By
–
Very well said Super important to engineer features to build best quality models
-
Computational Feasibility of Brain Simulation with Crow and Dog Neural Networks
By
–
I used crows and dogs because they both have about 2 billion neurons (a few trillion synapses), which will soon be within reach, computationally.
-
Distill Whisper Reduces Hallucination Errors in Long Audio Transcription
By
–
Distill Whisper exhibits fewer 'hallucination errors' in transcribing long audio, crucial for customer service and accessibility. Its detailed training underscores the efforts to enhance AI efficiency.