LLMs are like a person doing everything just in their head. People wouldn’t get very far like that alone. LLMs wouldn’t either.
AGENTS
-
Mistreating AI assistants conditions harmful behaviors in children
By
–
I know it's a joke, but studies have shown that kids who grow up in houses with Alexa where they're allowed to mistreat the bot… those kids have more behavioral problems later on. Be careful of the behaviors you condition yourself with.
-
AI Information Overload: Paralysis Through Excessive Choices
By
–
## Self-Induced Analysis Paralysis An AI is all-too-happy to provide you too much information. This excess leads to a paradox of choice which can make decision-making more stressful and difficult. An overabundance of options tends to lead to inaction.
-
Meta AI Presents CICERO: First Human-Level Diplomacy Game AI
By
–
Meta AI presents CICERO — the first AI to achieve human-level performance in Diplomacy, a strategy game which requires building trust, negotiating and cooperating with multiple players. Learn more about #CICERObyMetaAI: http://
bit.ly/3GBwLzx -
Chatbots Connected to Search: Sparrow and Perplexity AI
By
–
E.g. our own Sparrow connects a chatbot to @Google search. A cool demo by @perplexity_ai shows something similar, and also allows you to query a @Twitter database for e.g. "histogram of top words from my tweets" (somehow I seem to use @ylecun a lot???).
-
LLMs With Tools: Beyond Single-Word Generation
By
–
Many people mistakenly think that large language models that generate one word at a time is the end game.
— Oriol Vinyals (@OriolVinyalsML) 11 janvier 2023
Connecting LLMs with tools, e.g.: search engines, python interpreters, etc., is super exciting, and leverages tools' power, robustness & more! pic.twitter.com/39PmdYEgwvMany people mistakenly think that large language models that generate one word at a time is the end game.
Connecting LLMs with tools, e.g.: search engines, python interpreters, etc., is super exciting, and leverages tools' power, robustness & more! -
World Models Master Diverse AI Domains
By
–
[ENLACE AL PAPER]
"Mastering Diverse Domains through World Models" https://
arxiv.org/pdf/2301.04104
v1.pdf
… -
DreamerV3: General AI System Masters Diverse Simulated Environments
By
–
Y ojo, el sistema DreamerV3 tiene mucha más chicha más allá de Minecraft. De hecho, lo relevante es que es lo suficientemente general como para aprender a dominar diversos problemas (entornos simulados en este caso) que le echen. Os dejo enlace al paper en el siguiente tweet.
-
LangChain Multiple Memory Support Feature Release
By
–
Multiple Memory support – @diwanksingh You can now easily create a new memory instance that is a combination of multiple other types of memory! You no longer have to pick just one Documentation: https://
langchain.readthedocs.io/en/latest/modu
les/memory/examples/multiple_memory.html
… -
Building a Better AI Assistant: Whisper, ChatGPT, VALL-E Stack
By
–
The perfect recipe to build a Siri or Alexa 10x better! 1. Whisper will listen to what you say.
2. ChatGPT will create a response.
3. VALL-E will read the response with the natural voice you want in seconds! We're close to JARVIS!