Model techniques map
Techniquesoptimizationtraining runtime

implementation detail · filed under optimization

Minimal recomputation-graph extraction by backward traversal

Finding the minimal subgraph needed to recompute an annotated tensor by traversing the graph backward.

source
1
models
2
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

For each annotated tensor, it performs a backward traversal to identify the minimal subgraph required for its recomputation.

usedsoftware implementationin DeepSeek-V4DeepSeek

Filed alongside

Other methods under optimization :: training runtime.