Learn and Burn

Home
Archive
About
Measuring a model's understanding — starting with path-finding
[Paper: Evaluating the World Model Implicit in a Generative Model]
 • 
Unbox Research
Making LLMs scalable by replacing weights with learnable tokens
[Paper: Tokenformer: Rethinking Transformer Scaling with Tokenized Model Parameters]
 • 
Unbox Research
Image generation for infinite games
[Paper: Unbounded: A Generative Infinite Game of Character Life Simulation]
 • 
Unbox Research
Do LLMs rely on data contamination to solve math problems?
[Paper: GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models]
 • 
Unbox Research
1

October 2024

Running an LLM on a small customizable chip
[Paper: LlamaF: An Efficient Llama2 Architecture Accelerator on Embedded FPGAs]
 • 
Unbox Research
Better language models with negative attention
[Paper: Differential Transformer]
 • 
Unbox Research
A serious look at the future of AI medical advice
[Paper: A Preliminary Study of o1 in Medicine: Are We Closer to an AI Doctor?]
 • 
Unbox Research
LLMs have original, research-worthy ideas
[Paper: Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers]
 • 
Unbox Research

September 2024

OpenAI's o1 model
[Article: Learning to Reason with LLMs]
 • 
Unbox Research
The subgoals of attention units in LLMs
[Paper: Attention Heads of Large Language Models: A Survey]
 • 
Unbox Research
A model to parse body language
[Paper: Sapiens: Foundation Model for Human Vision Models]
 • 
Unbox Research
Stable diffusion can simulate video games
[Paper: Diffusion Models are Real-Time Game Engines]
 • 
Unbox Research
© 2024 Unbox Research
Privacy ∙ Terms ∙ Collection notice
Start WritingGet the app
Substack is the home for great culture