implementation detail · filed under model architecture
Loss-spike-triggered Anticipatory Routing
Activates Anticipatory Routing exclusively after a loss spike, alongside a short rollback.
- source
- 1
- models
- 2
- lab adopt it
- 1
- strongest
- optional
How sources treat it
One count per evidence span, weakest treatment to strongest.
optional 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
we introduced an automatic detection mechanism that triggers a short rollback and activates Anticipatory Routing exclusively when a loss spike occurs
optionaloptimizationin DeepSeek-V4DeepSeek
Filed alongside
Other methods under model architecture :: channel mixer :: mixture of experts :: expert routing.
Top-8 expert routingToken-level expert routingAnticipatory RoutingTop-4 expert routingDP-aware routingFixed Top-k routing with frozen biasHash routingKeep RoutingLatent-space routingRouted expertsToken-choice routingToken-choice routing with softplus gatingTop-6 expert routingTop-k expert routing with softmax over selected experts