“Cohere introduced a new AI model called Command-R. It’s a scalable generative model focused on enabling enterprises to deploy AI at a production scale with strong accuracy, low latency, longer context, and support for 10 key languages. https://twitter.com/rowancheung/status/1767410913622003916

Command-R: RAG at Production Scale – https://txt.cohere.com/command-r/

“A problem with RAG in practice: I asked a “talk to your documents” AI “what is important for me to know?” The RAG system pulled up three documents, for reasons that weren’t clear. Then the LLM justified the documents as being vital. Everything worked technically. Answer was BS” / X – https://twitter.com/emollick/status/1766864861928001617

“Zep AI (YC W24) is memory for LLMs. The most important thing in an age when LLM reasoning is rising and context window is exploding: an agent that remembers everything about you is now here https://twitter.com/garrytan/status/1767691500534788549

“On Preventing Hallucinations And Creating More Robust LLM Systems As we put more LLM apps in production, we find that preventing hallucinations is the biggest problem to overcome. Unlike humans, they don’t know what they don’t know. There are multiple ways to prevent… https://twitter.com/bindureddy/status/1767720610908778732

“few hit AI apps exist because LLMs are what i’d call very “strangely-shaped” tools most tools are built with a specific purpose in mind like a screwdriver or a car but LLMs were something we stumbled upon by predicting text and playing with RL – we didn’t design the shape of…” / X – https://twitter.com/nearcyan/status/1767163220807266514

Building Meta’s GenAI Infrastructure – Engineering at Meta – https://engineering.fb.com/2024/03/12/data-center-engineering/building-metas-genai-infrastructure/

Diffusion models from scratch – https://www.chenyang.co/diffusion.html

Paper page – Stealing Part of a Production Language Model – https://huggingface.co/papers/2403.06634

[2403.03893v1] From One to Many: Expanding the Scope of Toxicity Mitigation in Language Models – https://arxiv.org/abs/2403.03893v1

[2311.08364v1] Plum: Prompt Learning using Metaheuristic – https://arxiv.org/abs/2311.08364v1

[2403.09611] MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training – https://arxiv.org/abs/2403.09611

[2403.09629] Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking – https://arxiv.org/abs/2403.09629

Unlocking the conversion of Web Screenshots into HTML Code with the WebSight Dataset – https://huggingface.co/blog/websight

[2403.04652] Yi: Open Foundation Models by 01.AI – https://arxiv.org/abs/2403.04652

Generative SF: How to thrive in a crowded enterprise AI market – Lightspeed Venture Partners – https://lsvp.com/stories/generative-sf-how-to-thrive-in-a-crowded-enterprise-ai-market/

[2403.07974] LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code – https://arxiv.org/abs/2403.07974

[2403.07816] Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM – https://arxiv.org/abs/2403.07816

[2403.05812] Algorithmic progress in language models – https://arxiv.org/abs/2403.05812
Maximizing training throughput using PyTorch FSDP | PyTorch – https://pytorch.org/blog/maximizing-training/

Heads up! You’ve scrolled to the end of this category. There may have been just one or two links (above), so go back up and double check to be sure you didn’t quickly scroll down past it.

Be Sure To Read This Week’s Main Post:

This week’s executive overview and top links are here:

AI News #24: Week Ending 03/15/2024 with Executive Summary and Top 36 Links

The post you just read is an deep dive extension of my weekly newsletter, This Week In AI, an executive summary of the top things to know in AI. Each week, I create an accessible overview for laypeople to feel confident they are conversant with the week’s AI developments. I include a curated list of must-click links of the week, to offer everyone a hands-on opportunity to explore the most intriguing updates in artificial intelligence across various categories, including robotics, imagery, video, AR/VR, science, ethics, and more. Beyond the overview, I post these topic-based deeper dives (below). If you haven’t read this week’s overview, I recommend starting there.

Credits/Sources

Most of these weekly links come from just a few prolific oversharing sources. Please follow them, as they work hard to find the news each week and they make it a lot easier for me to compile.