Sitemap - 2023 - Learn and Burn
How to make more faithful images with training captions
An open source AI that proves theorems
A leap forward for text-to-3d generation
An elegant equivalence between LLMs and data compression
Generating better images without prompt engineering
LLMs with fast, infinite-length conversations
One-shot text-to-image face accuracy: Into the PhotoVerse
Finding weak points in AI game players
Neural models that build cities
"My voice is my passport:" AI speaks in your voice without training
Real-time scene rendering from a few photos
Pull up a chair, type your password ... I won't peek
LLMs can add hidden watermarks that survive human paraphrasing
AI + fMRI = reading minds (with music)
Finding the hidden phrases that jailbreak LLMs
Using gzip as a machine learning model
Paper: MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers
Paper: Tree of Thoughts: Deliberate Problem Solving with Large Language Models
DeepMind Paper: AlphaDev discovers faster sorting algorithms
Paper: Any-to-Any Generation via Composable Diffusion
Paper: LM vs LM: Detecting Factual Errors via Cross-Examination
Paper: LIMA: Less Is More for Alignment
We don't know why GPT works so well.
How Meta built a model that really can Segment Anything
This AI-populated matrix feels like Stardew Valley
I have seen the future of image-generating AI
How LLMs redefine the workforce
Bonjour, accent-free speech-to-speech translation!
A surprising, simple way to fight underfitting
LLMs can model other peoples' thoughts
Can you tell if this text was written by a human?
Oh, your AI only generates 2D images? How cute.