Revision History
NOSA (Native and Offloadable Sparse Attention) · 3 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Improve internal links (2026-10-13)
Version 3 · Oct 13, 2026, 10:18 AM
- Correction: the paper does not say vLLM or SGLang run out of memory; under its matched on-GPU-cache protocol the full-attention engines run a 16x smaller real batch (only the Hugging Face rows hit OOM). Minor: which studies used 20 sampled HELMET instances; HF repository creation vs modification dates; dated 'Preprint' comment; 16:1 head ratio applies to the 8B model; NOSA-8B compression window discrepancy with the SparDA paper flagged
Version 2 · Sep 16, 2026, 03:08 PM