BWhat changed
Stacked a cross-decoder over a self-decoder that produces reusable global key-value caches, reducing repeated layer-cache storage and enabling early-exit prefill.
Scientific scope
Stacked a cross-decoder over a self-decoder that produces reusable global key-value caches, reducing repeated layer-cache storage and enabling early-exit prefill. This architecture changes the model; it is not a universal drop-in exact acceleration for arbitrary pretrained Transformers. Needle retrieval and resource ratios are setup-specific, not general long-context reasoning guarantees.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. This architecture changes the model; it is not a universal drop-in exact acceleration for arbitrary pretrained Transformers. Needle retrieval and resource ratios are setup-specific, not general long-context reasoning guarantees.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- You Only Cache Once: Decoder-Decoder Architectures for Language Models
- Sun et al.: YOCO decoder-decoder shared key-value caching — official project source
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