8/ Meta-Transformer – performs unified learning across 12 modalities; supports tasks like fundamental perception (text, image, point cloud, audio, video), application (X-Ray, infrared, hyperspectral, & IMU), and data mining (graph, tabular, & time-series)
AI
-
Training Dense Retrievers for LLM In-Context Examples
By
–
9/ Retrieve In-Context Example for LLMs – presents a framework to iteratively train dense retrievers to identify high-quality in-context examples for LLMs.
-
Retentive Network: Foundation Architecture Improving LLM Training Efficiency
By
–
7/ Retentive Network – a foundation architecture for LLMs to improve training efficiency and inference; adapts retention mechanism for sequence modeling that support parallel representation, recurrent representations & chunkwise recurrent representation.
-
LLM Challenges: Brittleness, Evaluation, and Experimental Design
By
–
6/ Challenges & Application of LLMs – summarizes a comprehensive list of challenges when working with LLMs that range from brittle evaluations to prompt brittleness to a lack of robust experimental designs.
-
Top ML Papers Week: Llama 2, FlashAttention-2, Retentive Networks
By
–
Top ML Papers of the Week (July 17 – July 23): – Llama 2
– FlashAttention-2
– Meta-Transformer
– Retentive Network
– Challenges & Application of LLMs
– How is ChatGPT’s Behavior Changing Over Time?
… -
Meta releases Llama 2 foundation models from 7B to 70B parameters
By
–
1/ Llama 2 – a collection of pretrained foundational models and fine-tuned chat models ranging in scale from 7B to 70B; Llama 2-Chat is competitive on a range of tasks and shows strong results on safety and helpfulness.
-
Guide to Training Your Own Large Language Models
By
–
How to train your own Large Language Models https://
bit.ly/44wZq2d #AI #MachineLearning #DeepLearning #LLMs #DataScience -
Wittgenstein’s Philosophy Illuminates Large Language Models Theory
By
–
Indeed!! That Wittgenstein quote is exactly what I needed to read right now. Thanks! Extremely interesting in the context of LLMs
-
Defining Impressive: Creativity Over Technical Complexity in AI
By
–
Define impressive? It's a fun, properly articulated take on a meme that has taken the world by storm. I care less about how technically complex it was to pull off and how pixel perfect it is.
-
How Offices Evolved Over a Century of Work
By
–
How offices looked in each decade of the past hundred years https://
wapo.st/3NeRjRc via @washingtonpost #FutureofWork