File-backed weight adoption, which maps checkpoint files into shared memory and imports them into PyTorch or MLX as no-copy GPU tensors, removes the 'ingestion tax' on unified-memory systems at no measured throughput cost.
Gulavani, Alexey Tumanov, and Ramachandran Ramjee
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.OS 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The Ingestion Tax: Adopting File-Backed Weights in Tensor Frameworks
File-backed weight adoption, which maps checkpoint files into shared memory and imports them into PyTorch or MLX as no-copy GPU tensors, removes the 'ingestion tax' on unified-memory systems at no measured throughput cost.