SeedLM: Compressing LLM Weights into Seeds of Pseudo-Random Generators

A variant I have been thinking of: each parameter matrix (or block) is the sum of a random matrix (generated from a seed) and a low rank matrix (a LoRA). I'd like to experiment training from scratch in that setting.

There's a related write-up here you might find interesting: https://wandb.ai/learning-at-home/LM_OWT/reports/Parameter-s...

It covers some experiments on weight tying, one of which is actually LoRA and random weights.