iiinova — where your digital self keeps evolving

Reposted by Steven Fortney

Sung Kim @sungkim.bsky.social · 04/02/2025

ReAG - Reasoning Augmented Generation - No chunking, splitting vectorizing bs - Stateless, no vector DBs etc. - Supports any model (deepseek, o3-mini et al) - Reasoning traces - Metadata filtering - Typescript, Python support

2274 0

Reposted by Steven Fortney

Sung Kim @sungkim.bsky.social · 25/01/2025

DeepSeek is part of a quant trading firm, which probably operates out of the fanciest office imaginable, but why am I picturing this?

3392 0

Reposted by Steven Fortney

Sung Kim @sungkim.bsky.social · 27/12/2024

Apparently, you can run DeepSeek-V3 locally, provided that you have 8 M4 Pro 64GB Mac minis. ~5 tok/sec.

27911 0

Reposted by Steven Fortney

Ethan Mollick @emollick.bsky.social · 21/12/2024

I haven’t seen o3 yet & have been critical of benchmarks for AI but they did test against some of the hardest & best On GPQA, PhDs with access to the internet got 34% outside their specialty, up to 81% inside. o3 is 87%. Frontier Math went from the best AI at 2% to 25% Some other big ones, too

611116 0

Reposted by Steven Fortney

Sung Kim @sungkim.bsky.social · 18/12/2024

Genesis project A generative physics engine able to generate 4D dynamical worlds powered by a physics simulation platform designed for general-purpose robotics and physical AI applications.

33711 0

Reposted by Steven Fortney

Riku Murai @rmurai0610.bsky.social · 16/12/2024

Introducing MASt3R-SLAM, the first real-time monocular dense SLAM with MASt3R as a foundation. Easy to use like DUSt3R/MASt3R, from an uncalibrated RGB video it recovers accurate, globally consistent poses & a dense map. With @ericdexheimer.bsky.social* @ajdavison.bsky.social (*Equal Contribution)

510016 0

Reposted by Steven Fortney

Sung Kim @sungkim.bsky.social · 15/12/2024

They hypothesize that there exist key "forking tokens," such that re-sampling the system at those specific tokens, but not others, leads to very different outcomes. An example would be that a simple punctuation mark, or just a single token, can prompt an LLM to produce a different response.

2234 0

Reposted by Steven Fortney

Sung Kim @sungkim.bsky.social · 10/12/2024

Meta's SPDL: Faster AI model training with thread-based data loading. This framework-agnostic data loading solution utilizes multi-threading to achieve high-throughput in a regulator Python interpreter. Blog: ai.meta.com/blog/spdl-fa... Repo: github.com/facebookrese...

0112 0

Reposted by Steven Fortney

Sung Kim @sungkim.bsky.social · 09/12/2024

Jane Street, a quant trading firm has a very good YouTube channel. For comparison, DeepSeek is also a quant trading firm. They recently published a video on "Building Machine Learning Systems for a Trillion Trillion Floating Point Operations". Link: www.youtube.com/watch?v=139U...

youtube.com

Building Machine Learning Systems for a Trillion Trillion Floating Point Operations

YouTube video by Jane Street

1367 0

Reposted by Steven Fortney

Peyman Milanfar @docmilanfar.bsky.social · 08/12/2024

How are Kernel Smoothing in statistics, Data-Adaptive Filters in image processing, and Attention in Machine Learning related? My goal is not to argue who should get credit for what, but to show a progression of closely related ideas over time and across neighboring fields. 1/n

411221 0

Reposted by Steven Fortney

Khoa @khoavuumn.bsky.social · 08/12/2024

Real footage of a synthetic control model

14642118 0

Reposted by Steven Fortney

Jack Parker-Holder @jparkerholder.bsky.social · 04/12/2024

Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.

1523561 0

Reposted by Steven Fortney

Joey Politano🏳️‍🌈 @josephpolitano.bsky.social · 03/12/2024

America has three functional high capacity institutions left The Federal Reserve The Southern District of New York and The Delaware Court of Chancery

1053947 0

Reposted by Steven Fortney

Stacy Mitchell @stacyfmitchell.bsky.social · 01/12/2024

1. The conventional explanation for food deserts—that these places are too poor or too rural to generate enough spending on groceries, or too Black to overcome racist corporate redlining — fail to grapple with a key fact: food deserts didn’t used to exist. My new piece in The Atlantic.

theatlantic.com

The Mystery of Food Deserts

They didn’t materialize around the country for no reason. Something happened.

19655682487 0

Reposted by Steven Fortney

Jesse Bruhn @jessebruhn.bsky.social · 21/11/2024

My idea for Econ seminars: speakers can go for as long as they want and talk about whatever they want. But we change the norm so that the audience can leave whenever they want and it’s nbd. Let supply and demand for attention determine seminar length/structure, etc.

4241 0

Reposted by Steven Fortney

Eugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 20/11/2024

Being logged into wandb on your phone is a recipe for misery

9754 0

Reposted by Steven Fortney

Edward Grefenstette @egrefen.bsky.social · 19/11/2024

🌶️(?) take: Agents are somehow hot right because people realized that LLM output can be interpreted as a DSL which directs side effects in the world (e.g. tool calls) rather than just returning text in a chat/autocomplete sense. What are the open challenges? A 🧵... [1/11]

916531 0