DSpark Objectives

dspark_tiled_objective computes three losses for valid draft slots:

Loss	Purpose
Hard cross-entropy (CE)	Train on sampled draft labels.
Total variation (TV)	Compare the draft and target token distributions.
Confidence	Predict token acceptance when a confidence head is present.

The function processes draft tokens in bounded chunks. It keeps target logits and the target output head stop-gradient. The Markov head uses previous-token IDs from the target vocabulary. Loss labels use the selected draft-vocabulary order. The two vocabularies can have different sizes.

The function returns one DraftLossStats value. Its numerators combine the CE, TV, and confidence losses with the configured loss weights. Position weights scale both numerators and denominators. Reduce counts across data-parallel ranks before you call DraftLossStats.normalized.

build_dspark_provider attaches the Markov head and optional confidence head to a checkpointable draft body. PR #3711 defines the head shapes and checkpoint keys. Training-loop configuration and rollout execution are outside this change.