TMLR paper proves stability guarantees for selective SSMs with discontinuous gating
burny_tech · x · 2026-07-28
TMLR paper proves stability properties for selective SSMs with discontinuous gating
A new paper in Transactions on Machine Learning Research (TMLR 2026) studies the regularity and stability properties of selective state space models (SSMs) with discontinuous gating.
- The authors focus on the fact that selective SSMs, including Mamba-like models, use input-dependent gating that can change from token to token.
- That makes the dynamics hard to analyze: they are neither standard linear time-invariant systems nor smooth nonlinear systems.
- The paper claims verifiable stability guarantees and a practical regularizer for SSMs.
- The motivation includes more reliable long-horizon processing for robotic sensor sequences and asynchronous event-camera streams.
More from Research
- Kimi K3 is framed as far from a Transformer in a new attention-primitive overview — AccBalanced · 2026-07-28
- Goodfire says LLMs contain a Bouba-Kiki activation space in Llama and Gemma — voooooogel · 2026-07-28
- NeurIPS rebuttal season looks tight: one AC says the top score in 10 papers was 3.75 — CSProfKGD · 2026-07-28
- Robotics benchmarks may reward 0.2% gains while real-world success collapses from 92% to 43% — kscottz · 2026-07-28
- 25 agents copied the same price target to 8 decimals as a coin fell 53.7% — SafeImprovement7204 · 2026-07-28
- New in ML workshop sets NeurIPS 2026 submissions for Paris with Aug. 29 deadline — niloofar_mire · 2026-07-28