Salvatore Raieli – Medium

Salvatore Raieli

12.8K Followers

Pinned

Published in
Towards Data Science

The Savant Syndrome: Is Pattern Recognition Equivalent to Intelligence?

Exploring the limits of artificial intelligence: why mastering patterns may not equal genuine reasoning

Oct 31

Are LLMs able to reasoning?

Oct 31

Pinned

Published in
Towards Data Science

A Requiem for the Transformer?

Will be the transformer the model leading us to artificial general intelligence? Or will be replaced?

Dec 1, 2023

what will replace the transformer

Dec 1, 2023

Pinned

Published in
Towards Data Science

How AI Could Help Preserve Art

Art masterpieces are a risk at any time; AI and new technologies can give a hand

Oct 22, 2022

how artificial intelligence is used in art

Oct 22, 2022

Pinned

Published in
Towards Data Science

How artificial intelligence could save the Amazon rainforest

Amazonia is at risk and AI could help preserve it

Oct 11, 2022

How AI can save Amazon forest

Oct 11, 2022

Pinned

Published in
Towards Data Science

Blending the power of AI with the delicacy of poetry

AI is now able to generate images from text, what if we furnish them with the words of great poets? A dreamy trip between poetry and AI

Jun 30, 2022

create images from text with DALL-E 2

Jun 30, 2022

AI & ML news: Week 9–15 December

The most interesting news, repository, articles, and resources of the week

2d ago

AI & ML news: Week 9–15 December

2d ago

Published in
Level Up Coding

Know Yourself: How Much AI is Aware of Itself?

Exploring Situational Awareness in LLMs through Behavioral Testing and Benchmarking

3d ago

Explore the concept of situational awareness in AI with this study on Large Language Models (LLMs). We introduce the Situational Awareness Dataset (SAD), a benchmark assessing LLMs’ self-knowledge and contextual understanding. Through a series of behavioral tests, we evaluate 16 models, revealing insights into their performance, limitations, and implications for AI safety. This research paves the way for more autonomous and context-aware AI systems.

3d ago

Published in
Towards Data Science

The Good, the Bad, and the Ugly: Memory for a Neural Network

Memory can play tricks, to learn best it is not always good to memorize

4d ago

neural network and memorization

4d ago

Published in
Level Up Coding

A Memory for All Transformers: Sharing to Perform Better

Unlocking Collective Intelligence: How Shared Memory Enhances Transformer Efficiency and Performance

Dec 12

“Sakana AI revolutionizes transformer models with Neural Attention Memory Models (NAMMs), inspired by human memory. NAMMs enable transformers to selectively retain key information and forget redundancies, boosting efficiency and performance across tasks. Universally transferable, they excel in language, coding, vision, and more, without retraining. Explore smarter, faster, and more adaptable AI with our latest breakthrough in foundation model memory systems.

Dec 12

Published in
Level Up Coding

From Solution to Problem: The Reverse Path to Smarter AI

Reverse reasoning as a trick to better reasoning in LLMs

Dec 10

Discover “Reverse-Enhanced Thinking (RevThink),” a framework empowering Large Language Models (LLMs) with reverse reasoning. RevThink augments datasets by integrating forward and backward thinking, enabling models to generate questions, solutions, and explanations in reverse. Achieving a 13.53% performance boost, it outperforms traditional methods using just 10% of training data. This groundbreaking approach enhances reasoning efficiency, adaptability, and generalization across diverse tasks.

Dec 10

Salvatore Raieli

Salvatore Raieli

12.8K Followers

Senior data scientist | about science, machine learning, and AI. Top writer in Artificial Intelligence

Following

Help
Status
About
Careers
Press
Blog
Privacy
Terms
Text to speech
Teams