WRSWorld Research StateContribute
WORLD RESEARCH STATE

PICASO: composing precomputed SSM context states

Contribution grade

Retrieval-augmented recurrent inference · Human-designed research method or resource

Reference score45,000Provisional editorial estimate of this exact contribution; its grade is assessed separately.

What changed

Stored precomputed context states and transition factors, composed them algebraically and averaged compositions over permutations or cyclic rotations to reduce online context-processing costs.

Scientific scope

Stored precomputed context states and transition factors, composed them algebraically and averaged compositions over permutations or cyclic rotations to reduce online context-processing costs. Concatenation identity is exact for the stated single-layer input-dependent linear dynamics, not general stacked nonlinear SSMs. Full-permutation formulas require commuting transitions; cyclic averaging is not full permutation invariance. Multilayer use is approximate; offline preprocessing/storage and retrieval costs remain. The strongest quality/speed comparisons are source-reported on selected datasets.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Concatenation identity is exact for the stated single-layer input-dependent linear dynamics, not general stacked nonlinear SSMs. Full-permutation formulas require commuting transitions; cyclic averaging is not full permutation invariance. Multilayer use is approximate; offline preprocessing/storage and retrieval costs remain. The strongest quality/speed comparisons are source-reported on selected datasets.

Contributors

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction