Multiverse Computing's pruning method beats a baseline by nearly 23 MMLU points
The method searches interacting block-removal choices instead of ranking layers one at a time, with its biggest gains appearing under severe compression.
By RuntimeWire Staff · Published
Primary source: Hugging Face
Why it matters
If the results hold across independent evaluations, model builders gain a cheaper way to search aggressive pruning configurations, while Multiverse Computing gains a bridge from quantum optimization to a large, immediate AI market.
Roman Orus, the quantum-information physicist who co-founded Multiverse Computing, and a team of researchers have recast one of the bluntest ways to shrink a large language model as a problem from statistical physics. In a technical account published on Hugging Face on September 21st, the researchers described a method for choosing whole transformer blocks to delete by modeling the choices as an Ising glass.…