A curated collection of Weight-Space Programming and weight-centric research, covering weight understanding, generation, editing, merging, transfer, and continual learning.
-
Updated
Nov 20, 2026 - HTML
A curated collection of Weight-Space Programming and weight-centric research, covering weight understanding, generation, editing, merging, transfer, and continual learning.
Research: can a token's weights be generated from a 64-bit code rather than looked up from a global tensor? Binary activations, shared low-rank basis, 41.6M-param LM.
Make your own small AI that runs on a laptop: teach it your documents or describe how it should answer. Honest report cards, fixes by editing weights, one-file export for Ollama.
CNN with DNA-generated filters
Experimental PyTorch utility for coordinate-based neural-weight generation.
DNA research framework
Experimental PyTorch Fourier-feature generator for coordinate-based weight matrices.
DNA compression + HELIX memory
Production-ready DNA generator
GAN with compressed generator
Lightweight DNA for small models
Sparse matrix generation with DNA
MLP with DNA compression
Quantized DNA weight generation
RNN with DNA compression
Transformer with DNA-generated weights
To associate your repository with the weight-generation topic, visit your repo's landing page and select "manage topics."