April
Papers
- CoLT5: Faster Long-Range Transformers with Conditional Computation
- ZeRO: Memory Optimizations Toward Training Trillion Parameter Models
- Token Merging for Fast Stable Diffusion
- LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention
- Segment Anything
- Evolving Neural Networks through Augmenting Topologies (NEAT)
- DINOv2: Learning Robust Visual Features without Supervision
Links
- https://flatassembler.net/ Flat assembler
- https://exploit.education/phoenix/ Resource on binary exploitation
- https://pwn.college/ Resource for learning CTFs
- https://developer.chrome.com/blog/webgpu-release/ Chrome ships WebGPU
- https://jott.live/markdown/webgpu_safari Using WebGPT to implement matrix multiplication
- https://llvm.org/docs/tutorial/MyFirstLanguageFrontend/index.html Resource for learning and building a complier
- https://craftinginterpreters.com/scanning.html Resource for building a compiler
- https://n0.lol/ n0.lol
- https://astexplorer.net/ Nice abstract syntax tree visualiser, works with multiple languages
- https://guyinatuxedo.github.io/index.html Introduction course to binary exploitation
- https://webgpufundamentals.org/ WebGPU fundamentals
- https://github.com/antonio-morales/Fuzzing101 Resources on fuzzing
- https://github.com/seal9055/resources/ Resources for cyber security and exploitation
- https://0dayfans.com/ List of published vulnerabilities
- https://wumb0.in/ Interesting blog on reversing and exploitation
- https://oimo.io/works/life/ Game of life in the game of life
- https://en.wikipedia.org/wiki/One-instruction_set_computer One instruction set computer
- https://blog.eleuther.ai/transformer-math/ Information on math related to the computation and memory usage for transformers
- https://transformer-circuits.pub/2022/mech-interp-essay/index.html Mechanistic interpretability, variables, and the importance of interpretable bases
- http://www.paulgraham.com/nerds.html Why nerds are unpopular
- https://distill.pub/2020/understanding-rl-vision/ Mechanistic interpretability for reinforcement learning vision
- https://distill.pub/2019/activation-atlas/ Exploring neural networks with activation atlases
- https://worldmodels.github.io/ Recurrent world models facilitate policty evolution
- https://mlu-explain.github.io/double-descent/ Double descent in deep learning