TLDR Index
TLDR AI

TLDR AI 2024-06-05

6 sections19 entrieslinks blockcomplete parse

Parse notes

  • Subscription/account footer truncated before parsefooter_truncated

01GENERAL

Sponsored

Email is one of the main ways companies communicate with customers — but very few companies know how to parse unstructured email text into useful data [4].

nylas.comSponsor

02HEADLINES & LAUNCHES

A RIGHT TO WARN ABOUT ADVANCED AI

A group of current and former AI employees is calling for advanced AI companies to commit to principles that ensure transparency and protection for employees who raise risk-related concerns. They highlight the need for companies to avoid enforcing non-disparagement agreements, facilitate anonymous reporting processes, support open criticism, and prevent retaliation against whistleblowers.

righttowarn.ai3 min read

WILL SCALING SOLVE ROBOTICS?

Over 900 people attended last year's Conference on Robot Learning, which featured 11 workshops and almost 200 accepted papers. One of the largest debates at the event was whether training a large neural network on a very large data set was a feasible way to solve robotics. This post presents the different sides of the argument to deepen people's understanding of the debate. Scaling has worked for other similar fields. However, it is impractical as there isn't much robotics data available and there's no clear way to get it. Even if scaling works as well as in other fields, it likely still won't solve robotics.

spectrum.ieee.org15 min read

THE OPAQUE INVESTMENT EMPIRE MAKING OPENAI'S SAM ALTMAN RICH

Sam Altman is one of Silicon Valley's most prolific and aggressive individual investors. He manages an investment empire with holdings worth at least $2.8 billion as of early this year. Much of the portfolio isn't widely known. This article takes readers through what's known about Altman's investments.

wallstreetsights.com15 min read

03RESEARCH & INNOVATION

MMLU PRO

MMLU is a common benchmark for reasoning tasks. It is often considered to be both the gold standard and something models have overfit. MMLU Pro is a new, harder, and cleaner benchmark for measuring language model reasoning.

arxiv.org26 min read

TREE DIFFUSION: DIFFUSION MODELS FOR CODE

Fantastic diffusion paper that diffuses code for images. It can directly make edits as part of the diffusion process. It is slow, but can be combined easily with search to dramatically improve reasoning ability.

tree-diffusion.github.io18 min read

JAILBREAKING OF LARGE LANGUAGE MODELS

Researchers have introduced improved methods for optimization-based jailbreaking of large language models, building on the Greedy Coordinate Gradient (GCG) attack.

arxiv.org16 min read

04ENGINEERING & RESOURCES

Sponsored

WHAT IS RAG? — A NEO4J BLOG

A chatbot can sound human AND provide detailed, precise answers. Check out this blog from Neo4j about how retrieval-augmented generation (RAG) IMPROVES YOUR LLM'S RESPONSES [11]. Learn about use cases where RAG made a huge difference in making a chatbot's answers

neo4j.comSponsor

OMOST IMAGE SYNTHESIS

From the creator of ControlNet, Omost is a way to gain control over your image generation. It first rewrites prompts as a set of descriptive code. Then it uses that to render the final image. Importantly, you can edit the code before or after generation to slightly change the model output.

github.comGitHub repository

IMPROVED VIDEO SUPER RESOLUTION

Researchers have developed a training-free video interpolation method for generative video diffusion models. This new approach, compatible with various models, enhances frame rates without the need for extensive training or large datasets.

ssyang2020.github.io4 min read

IMAGE COMPRESSION

Control-GIC is a new framework for generative image compression that allows fine-grained bitrate adjustment while maintaining high-quality results.

github.comGitHub repository

05MISCELLANEOUS

LLM INFERENCE SPEED OF LIGHT

Using theoretical speed of light modeling as grounding is really important for problems where the amount of computation and memory access is known a priori as it helps validate the quality of implementations and predict the impact of architectural changes.

zeux.io13 min read

PLENTIFUL, HIGH-PAYING JOBS IN THE AGE OF AI

It's possible that many of the jobs that humans do today will continue to be done by humans indefinitely, no matter how much better AIs are at those jobs, due to comparative advantage.

noahpinion.blog20 min read