Adam Chess's Avatar

Adam Chess

@arsonade.bsky.social

M.A. philosophy, MPH Healthcare Data Scientist. He/Him. Philosophy, public health, stats, data science, scifi, art, TTRPGs, game design, and comics. I mostly just repost when I learn something

651 Followers  |  3,361 Following  |  255 Posts  |  Joined: 25.08.2023  |  2.0334

Latest posts by arsonade.bsky.social on Bluesky

I came to Minneapolis to report on what's going on, and one of the main questions I showed up with is "just what is the scale of the resistance?" After all, we're all used to the news calling Portland a "war zone" or whatever when it's just some protests in one part of town.

22.01.2026 03:58 โ€” ๐Ÿ‘ 15512    ๐Ÿ” 5791    ๐Ÿ’ฌ 141    ๐Ÿ“Œ 1563
Post image

We introduce epiplexity, a new measure of information that provides a foundation for how to select, generate, or transform data for learning systems. We have been working on this for almost 2 years, and I cannot contain my excitement! arxiv.org/abs/2601.03220 1/7

07.01.2026 17:27 โ€” ๐Ÿ‘ 144    ๐Ÿ” 34    ๐Ÿ’ฌ 9    ๐Ÿ“Œ 9

Yep. This has been my approach so far

09.12.2025 20:43 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Preview
Titans + MIRAS: Helping AI have long-term memory

TITANS & MIRAS: real continual learning

MIRAS = a unifying theory of transformers (attention) and state space models (SSM, e.g. Mamba, RNNs)

TITANS = an optimal MIRAS implementation thatโ€™s โ€œhalfway betweenโ€ SSM & transformer with a CL memory module

letโ€™s dive in!

research.google/blog/titans-...

06.12.2025 16:04 โ€” ๐Ÿ‘ 42    ๐Ÿ” 11    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 1

I use R at work for lots of things that aren't really analysis, but within that subset I'd say it's probably 70% causal vs 30% predictive

03.12.2025 14:17 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

And turkey sucks too, but try the deep fried bird one day (with proper precautions) and you'll never want to go back

Thanksgiving food is easy to make but hard to master

27.11.2025 04:35 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

You're right and wrong. Stuffing sucks, but Ive seen people elevate it to delicious specifically as a reaction to that. Cranberry was a terrible tradition until my mom found a recipe with cranberry + jello + cream cheese that is now the thing we look forward to. Pumpkin pie -> sweet potato

27.11.2025 04:33 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

no wonder g_d was pissed when they ate that fruit

24.11.2025 06:03 โ€” ๐Ÿ‘ 104    ๐Ÿ” 20    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 1
Preview
Public Health Was a Place for Warriors Once. It Needs to Be Again. It's time for us to reconnect with the radical, system-changing spirit that was once at the heart of our field.

Itโ€™s time for us to reconnect with the radical, system-changing spirit that was once at the heart of our field. #publichealth #episky #medsky #activism www.thenation.com/article/acti...

06.11.2025 14:40 โ€” ๐Ÿ‘ 260    ๐Ÿ” 78    ๐Ÿ’ฌ 6    ๐Ÿ“Œ 4
Preview
Inside a 1,200 Player RPG A closer look at Over/Under, the massive play-by-post game

Want a peek inside a 1,200 player RPG?

Read out interview with @samsorensen.bsky.social about OVER/UNDER, the massive play-by-post game that's a part of Mothership Month ( @mothership.bsky.social )

manysidednewsletter.substack.com/p/inside-a-1...

04.11.2025 17:19 โ€” ๐Ÿ‘ 45    ๐Ÿ” 21    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 2
Video thumbnail

Do AI agents ask good questions? We built โ€œCollaborative Battleshipโ€ to find outโ€”and discovered that weaker LMs + Bayesian inference can beat GPT-5 at 1% of the cost.

Paper, code & demos: gabegrand.github.io/battleship

Here's what we learned about building rational information-seeking agents... ๐Ÿงต๐Ÿ”ฝ

27.10.2025 19:17 โ€” ๐Ÿ‘ 24    ๐Ÿ” 11    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 2
Preview
Protein Powders and Shakes Contain High Levels of Lead - Consumer Reports CR tests of 23 popular protein powders and shakes found that most contain high levels of lead.

my latest investigation for @consumerreports.org is based on months of reporting and 60+ lab tests of leading protein supplements

we found that most protein powders and shakes have more lead in one serving than our experts say is safe to have in a day (๐Ÿงต)

www.consumerreports.org/lead/protein...

14.10.2025 16:37 โ€” ๐Ÿ‘ 6143    ๐Ÿ” 3259    ๐Ÿ’ฌ 285    ๐Ÿ“Œ 1491

The core was essentially taking shame as the means to gain supernal power (like how vamps drain vitae). It's hard to seperate character embarrassment from OOC - you wind up feeling it. You need that shame to be useful, the more genuine the more effective, and the more thoughts of revenge creep in

13.10.2025 23:00 โ€” ๐Ÿ‘ 3    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Preview
Hunchback: The Lurching Main Page > New World of Darkness > Hunchback: The Lurching Right off the bat, letโ€™s get one thing settled. This is NOT a game about persons with any type of congenital birth defect or malformation. I...

This wierd fuckin broken thing.

Somehow we Homebrewed the rules enough to make it work and the result was somehow this amazing synthesis of role play and system that made me realize how those things could feed into each other. Also how a game that wasn't 'fun' could still be very much fun.

13.10.2025 22:38 โ€” ๐Ÿ‘ 1    ๐Ÿ” 1    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Over the past year, my lab has been working on fleshing out theory + applications of the Platonic Representation Hypothesis.

Today I want to share two new works on this topic:

Eliciting higher alignment: arxiv.org/abs/2510.02425
Unpaired learning of unified reps: arxiv.org/abs/2510.08492

1/9

10.10.2025 22:13 โ€” ๐Ÿ‘ 132    ๐Ÿ” 32    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 5
Post image

if you're curious about the architecture and mechanics of LLMs, this site has a really excellent explorable interactive visualization. it helps build intuition for how massive these models are, what 'interpretability' means, and the complexity involved here

bbycroft.net/llm

06.10.2025 17:51 โ€” ๐Ÿ‘ 161    ๐Ÿ” 24    ๐Ÿ’ฌ 10    ๐Ÿ“Œ 2
Preview
Democrats Must Embrace War Mindset We have exited a long period of peace and entered into a struggle not unlike war, in which all the old certainties have gone malleable and the future of the republic is at stake.

โ€œ[D]ooming is itself a liberation from the burden of choice. If everything is ruined forever, if your allies have already forsaken you, if the battle is already lost, you aren't responsible for your choices. They can't affect the outcome. You're free.โ€

04.10.2025 15:22 โ€” ๐Ÿ‘ 12    ๐Ÿ” 8    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
The 2500th, 250th, 50th, and 25th largest model families on Hugging Face. They show varying numbers of generations (between 3 and 8) and different edge types, including adapters, finetunes, merges, and quantizations.

The 2500th, 250th, 50th, and 25th largest model families on Hugging Face. They show varying numbers of generations (between 3 and 8) and different edge types, including adapters, finetunes, merges, and quantizations.

In a new paper with @didaoh and Jon Kleinberg, we mapped the family trees of 1.86 million AI models on Hugging Face โ€” the largest open-model ecosystem in the world.

AI evolution looks kind of like biology, but with some strange twists. ๐Ÿงฌ๐Ÿค–

14.08.2025 15:06 โ€” ๐Ÿ‘ 31    ๐Ÿ” 9    ๐Ÿ’ฌ 4    ๐Ÿ“Œ 2
Preview
First known wild โ€˜grue jayโ€™ hybrid spotted in Texas Green and blue jays are crossing paths as temperatures rise.

Oh. My. God.

www.popsci.com/environment/...

23.09.2025 14:53 โ€” ๐Ÿ‘ 165    ๐Ÿ” 30    ๐Ÿ’ฌ 9    ๐Ÿ“Œ 8

I've always wondered about this too and would prefer the alternative most of the time, but I do feel like for many systems you need at least a few sessions before everyone is familiar with it enough to let it shine. Also people like gradual progression

22.09.2025 20:39 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

okay AI controversy aside, just talking as a nerd for a second: this is such a cool paper

21.09.2025 16:07 โ€” ๐Ÿ‘ 792    ๐Ÿ” 178    ๐Ÿ’ฌ 16    ๐Ÿ“Œ 4

LLMs becoming superhuman at all games & competitions but not even normal-expert good at anything that's not a game or competition continues to raise deep questions about whether games & competitions are a secret natural kind

18.09.2025 00:08 โ€” ๐Ÿ‘ 60    ๐Ÿ” 5    ๐Ÿ’ฌ 5    ๐Ÿ“Œ 0

Smoking Kills

10.09.2025 16:16 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
2 Minute Deep Acid in Strudel (from scratch)
YouTube video by Switch Angel 2 Minute Deep Acid in Strudel (from scratch)

youtu.be/HkgV_-nJOuE?...

See this one yet? What's one more hobby right?

31.08.2025 03:21 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Solar luminosity around 3.5 billion years ago was only about 70% of what it is right now.

Which means that when life arose on earth, it was outside the sun's habitability zone: liquid water should not have been able to exist.

Faint Young Sun Paradox

07.04.2024 23:40 โ€” ๐Ÿ‘ 60    ๐Ÿ” 8    ๐Ÿ’ฌ 3    ๐Ÿ“Œ 3

Exactly. To some extent thats always been exaggerated for plot, but not that much, and we lived in a world where those stakes were believable just last year. Some Black Mirror come to mind. SCP foundation and Co. Random but what prompted this for me was the cover art for Jurassic 5's new album

26.08.2025 15:40 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

It seems clear that the current historical moment will have a major impact on culture. Even recent art feels anachronistic. Movies involving govt/cops more obviously, but also just the undertones of social stability. It feels like it meant something different to make music last year. Any art really.

26.08.2025 14:29 โ€” ๐Ÿ‘ 3    ๐Ÿ” 0    ๐Ÿ’ฌ 2    ๐Ÿ“Œ 0

I really enjoyed this. Even just reading it. Well done

26.08.2025 14:10 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

I'm pretty sure it's more like 'the left likes this guy and wants him free, therefore we will do everything in our power to cause him pain'. They don't care if people think it was a mistake. The point is to show that resistance will be met with escalation and that the inciting cause doesn't matter

25.08.2025 16:03 โ€” ๐Ÿ‘ 5    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

@arsonade is following 20 prominent accounts