Jonathan Cheng
Jonathan Cheng
@jonathancheng.bsky.social
English Literature PhD turned ML Researcher. I convert cold brews into noisy experiments. Foundation Models @ Apple NYC 🏳️🌈
4.8K followers1.9K following1.3K posts
Jonathan Cheng reposted
Replying to @mariaa.bsky.social
Jonathan Cheng reposted
Replying to @kjhealy.co
Which all just reeks of what happens when a status order gets upended by some new technique. Basically everyone who is under threat runs around shouting “Never mind what I appeared to value before, what I *really* value is this incredibly intangible thing I possess that also preserves my status”.
Jonathan Cheng reposted
New from our lab! #COLM2026 When people generate stories, they don't just write one prompt. Instead, they explore narrative space via branching edits 🌱 We reconstruct 24k of these edit trees 🌳 from chat logs and map edit types, story formats, how they relate to tree depth, and more!
This paper and its predecessor have just been a joy to read.
On-Policy Distillation is surprisingly data-efficient, with a single query recovering 72% of the full-data gain because rollouts already cover 71.5% of the relevant state space. So the bottleneck is how slowly the student absorbs dense token-level teacher supervision www.alphaxiv.org/abs/2609.04172
It’s been a long week, but my god. Silent hill returning to its New England-Rust Belt Gothic roots. Still not sure why Japanese renderings of Centralia was such a thrill for my pre-teen self 😂 youtu.be/df9v2l1kEx8?...
The passage describing the wall of contorted, angry statues was foreshadowing…of bsky’s displeasure towards someone not liking Piranesi. So it’s more like a Hall. In the southwest wing, in the year the albatrosses arrived.
This is the way that Bluesky is a bubble. Everyone loves Piranesi. Everyone goes crazy when someone new reads Piranesi.
Jonathan Cheng reposted
Jonathan Cheng reposted
Jonathan Cheng reposted
Jonathan Cheng reposted
New paper! w/ @cocoscilab.bsky.social🧵Can large language models reason flexibly, or have they learned what reasoning looks like? We introduce a new paradigm to test this question—the riddle riddle—and find that humans and LLMs show opposite patterns of performance. 📜
Jonathan Cheng reposted
Jonathan Cheng reposted
Jonathan Cheng reposted
The Pudding's early history of U.S. restaurant dining: don't miss the interactive map of 5,000 historical menus pulled from the NYPL's Buttolph collection https://pudding.cool/2026/06/m...
Wish I knew Persona 4 was going to feature so prominently!
Excited to share this. @neel2112.bsky.social, @mariaa.bsky.social, and I analyzed 500K anonymous ChatGPT convos (shared w/ consent from WildChat) to see if people were generating fiction. We found tons of stories, fanfiction & erotica. Many users iterated on the same stories for days and weeks.
Jonathan Cheng reposted
Apparently, there is a cafe in Shenzen’s biotech hub that gives you free coffee for a month if you publish in @nature.com or @science.org. ;-)
I think my boss has correctly identified that, despite my best efforts, I try to take work seriously. And this can lead to all sorts of anguish/frustration. She sent me this lovely blog post — it even includes a lovely reference to Sima Yi. www.linkedin.com/pulse/half-y...
Through my partner’s wish for aesthetic explore > exploit. We watched Felidae. en.wikipedia.org/wiki/Felidae... I would describe this German film as: made by an unhappy man who likes zdzisław beksiński paintings and decided Lion King would be better with more commentary on eugenics. Meow/10
Jonathan Cheng reposted
Everyone should read this, just to see how they implemented a perceptron (a very basic neural network) in Age of Empires II arxiv.org/abs/2605.31514 The artificial neurons pass activation energy in the form of goats.