Revision History
State space model (deep learning) · 10 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.
- Update: added a video.
Version 10 · Nov 1, 2026, 08:18 AM
- Correction: fixed S4 discretization (bilinear, not ZOH) and kernel complexity, SaShiMi described as a benchmark (it is a model), Mamba-3 authors/year and state-size claim (MIMO variant), Mitra et al. speedups missing their consumer/embedded-GPU conditions, unsupported 'ICL mechanisms' claim, Mamba-2 'simplified gating', Nemotron-H 47B/1M-context wording, RWKV-X date, Griffin/RecurrentGemma described as SSM hybrids, unverified Nature Communications claim; added selective-copying/induction-heads results, S6 parameterization, Mamba-2 block changes, SSM vs linear-attention boundary, recall evidence (Zoology, Repeat After Me), and a sourced 2024-2026 hybrid model table.
Version 9 · Oct 31, 2026, 03:34 AM
- Start the article with its lead paragraph: moved the first paragraph above the "What is a state space model?" heading. No text changed.
Version 8 · Oct 29, 2026, 09:24 AM
- Improve internal links (2026-10-13)
Version 7 · Oct 13, 2026, 10:07 AM
- Added 6 contextual internal links
Version 6 · Jul 23, 2026, 03:04 PM