BWhat changed
Trained multimodal understanding and generation under a common next-group-of-tokens objective, with elastic depth, expert capacity and routing sparsity to support multiple deployment budgets from one run.
Scientific scope
Trained multimodal understanding and generation under a common next-group-of-tokens objective, with elastic depth, expert capacity and routing sparsity to support multiple deployment budgets from one run. Distinct from ERNIE4.5's heterogeneous expert sharing. Elastic/supernetwork training and autoregression have prior art; no first-ever or universal optimality claim is imported. Submodel quality and post-training stability are source-reported.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Distinct from ERNIE4.5's heterogeneous expert sharing. Elastic/supernetwork training and autoregression have prior art; no first-ever or universal optimality claim is imported. Submodel quality and post-training stability are source-reported.
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.
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