3. Minimize cross-entropy loss between predicted and actual next tokens. (This is equivalent to trying to minimize encoding length.) ViT don't have quantization baked into the algorithm, but NNs are being moved towards quantization in general. ----- I think it's pretty interesting how closely related generation and compression are. ClosedAI's Sora[^1] model uses a denoising vision transformer, and is the... - Source: Hacker News / 26 days ago

Did you miss the fish?[0] You should see the error in first viewing What about the woman with glasses? Her face literally "jumps"[1] Same with this guy's hands[2] Interesting, we notice that [1] has "sora" in the name though I think it is a reference to the main image on sora[3] Not sure if the gallery is weird to anyone else, but it doesn't exactly show new images and the position indicator is wonky. The thing... - Source: Hacker News / about 1 month ago

Been there myself! For one you may not have heard of, check out Recollectr. Source: about 1 year ago

Perhaps I'm biased as the creator, but as someone who also has ADHD, Obsidian seems overwhelming. I'm the creator of Recollectr which has a lot of the same functionality built into a much more streamlined UI. Source: almost 2 years ago

I wasn't satisfied with Standard Notes either; so how about Recollectr? Source: almost 2 years ago

I've spent the past few years making Recollectr, a desktop-first note-taking app to help you, "Record and recall without breaking your flow.". Source: almost 2 years ago

We just pushed the new version (3.16.199) out via the auto-updater; it's also available on https://recollectr.io too. Source: almost 2 years ago