LLMS

→ View original post on X — @ken_goldberg

Dismantling Language Barriers to Advance Global Communication and Collaboration

By

@ken_goldberg

–

23 June 2023 5h03

Dismantling the Tower of Babel would significantly advance human communication, education, and collaboration. Of course there will be unintended consequences but the upside seems undeniable. https://t.co/9SwbMJ85gh
— Ken Goldberg (@Ken_Goldberg) 23 juin 2023

Dismantling the Tower of Babel would significantly advance human communication, education, and collaboration. Of course there will be unintended consequences but the upside seems undeniable.

Quantizable Transformers: Removing Outliers from Attention Heads

By

–

23 June 2023 3h13

Quantizable Transformers: Removing Outliers by Helping Attention Heads Do Nothing paper page: https://
huggingface.co/papers/2306.12
929
… Transformer models have been widely adopted in various domains over the last years, and especially large language models have advanced the field of AI

Google Introduces AudioPaLM: Speech-Capable Large Language Model

By

–

23 June 2023 3h05

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen

paper page: https://t.co/uLZwULDc94

introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al.,… pic.twitter.com/85fyv2B25R
— AK (@_akhaliq) 23 juin 2023

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: https://
huggingface.co/papers/2306.12
925
… introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al.,

Word Models to World Models: Natural Language to Probabilistic Thought

By

–

23 June 2023 3h00

From Word Models to World Models: Translating from Natural Language to the Probabilistic Language of Thought paper page: https://
huggingface.co/papers/2306.12
672
… How does language inform our downstream thinking? In particular, how do humans make meaning from language — and how can we leverage

Deep Language Networks: Joint Prompt Training of Stacked LLMs

By

–

23 June 2023 2h53

Deep Language Networks: Joint Prompt Training of Stacked LLMs using Variational Inference paper page: https://
huggingface.co/papers/2306.12
509
… We view large language models (LLMs) as stochastic language layers in a network, where the learnable parameters are the natural language prompts at each

→ View original post on X — @rasbt

No Prominent Imitation Model for Coding Yet

By

@rasbt

–

23 June 2023 1h31

No worries! On that note, I don't think there is an imitation model for coding, yet. At least nothing prominent as far as I know.

→ View original post on X — @replit

Steve’s Fast-Track Tech Stack: FlowWise, Pinecone, GPT-3.5

By

@replit

–

23 June 2023 0h59

Steve chose his tech stack with speed to market in mind. He used FlowWise and a popular vector DB, Pinecone, to power the backend. The LLM used was GPT-3.5.

→ View original post on X — @replit

Building Finance AI Chatbots: Steve’s Week-Long Launch

By

@replit

–

23 June 2023 0h59

AI chatbots are one of the most popular categories on Replit Bounties. These projects are straightforward
– Corpus of data
– Embedded vector DB
– LLM API But they can vary based on the industry. Here's how one entrepreneur, Steve, launched his own finance AI bot in a week