A Watermark for LLMs – a watermarking framework for proprietary language models. 4 of 11
RESEARCH
-
H3: Bridging Performance Gap Between State Space Models and Attention
By
–
H3 – an approach to reduce the gap, in terms of performance and hardware utilization, between state space models and attention for language modeling. 3 of 11
-

AI Glossary: Prediction, Inference, and Generalization Explained
By
–
In this @Cognilytica #AIToday #podcast AI Glossary Series episode, 'Prediction, Inference, and Generalization' hosts @rschmelzer & @kath0134 define and discuss these terms and how they fit into the overall picture of #AI. Full episode: https://
cognilytica.com/ai-today-podca
st-ai-glossary-series-prediction-inference-and-generalization/?utm_source=dlvr.it&utm_medium=twitter
…
#tech #data #ML -

Visual Machine Learning Concepts Learning Website
By
–
Visual explanations of core machine learning concepts. A brilliant website for everyone looking to learn important machine learning concepts through visuals in a fun way. Check this: http://
mlu-explain.github.io -
spaCy Floret Vectors: New Approach for Transformer Models
By
–
Here's how we do it in @spacy_io : https://
explosion.ai/blog/floret-ve
ctors
… We're interested to try this trick in transformer models. We haven't really tried that yet. -
LogSumExp Function as Universal Solution for Numerical Stability
By
–
Yup… logsumexp is our friend for everything else.
-

Multimodal API from Aleph Alpha provides text completions with image context
By
–



This multimodal API from @Aleph__Alpha provides text prompt completions using images as context, like Flamingo. In each screenshot, the bold text is the prompt, and the non-bold text is AI-generated:
-

PhD Researcher in Domain Adaptation Seeking Scientist Position
By
–
I am on the job market for researcher/scientist position this year. I will stay in Tokyo for a research internship until end of May. My research during PhD focuses on model adaptation under domain and category shift. More details can be found in: https://
shiqiyang.xyz -
Building a Transformer with simplified layers in nanoGPT
By
–
Yep, that's the one! (as @Thom_Wolf linked earlier too). I'd expect it's possible to build a Transformer with that kind of layer alone, would look much more pleasing. Will see if I can prototype in nanoGPT.
-
Defining True AI: Current Definition Versus Reality
By
–
Por entender tu punto de vista entonces ¿cuál es la definición correcta de inteligencia artificial y en qué difiere a lo que actualmente se llama "inteligencia artificial"?
