02_第１部：記号接地していないchatGPT 2/6

[Audio notes] Scaling Laws for Neural Language Models

Flash Attentions (KOR) #attention #transformer

New Breakthrough on a 90-year-old Telephone Question

[Active learning helps pretrained models learn the intended task] 설명

Time Series Analysis with Stationarity

ZeroQuant Series - Jinsol Kim at Neubla (KOR) #DeepSpeed #Quantization #LLM #Transformer #ZeroQuant

[Audio notes] InstructGPT - Training language models to follow instructions with human feedback.

LAION-5B: Dataset for training image-text models

Self Supervised Anomaly Detection, Integration of Time Series Downsampling and Convolution

Efficient LLMs with more data & instruct based LLMs (KOR) #llm #gpt #chatgpt

Mixture-of-Experts with Expert Choice Routing

You can grow new brain cells. Here's how | Sandrine Thuret | TED

LLM UNDERSTANDING: 35. Irina RISH "Emergent Behaviors in Foundational Models"

Flamingo: Visual Language Model for Few-Shot Learning

Model Selection in Domain Generalization

So... why did AI take off now?

[DS Interface] DENSE: Data-Free One-Shot Federated Learning

Day 8 - A. Efros: Lessons from Computer Vision: We are (still!) not giving Data enough credit

The U-Net Model