R2 (visit ordered)
LLaMA-large HMTE decoder; within-visit events kept in observed order (standard cross-entropy).
Mass-corrected, modality-balanced backbone each source keeps its top targets in every modality; no connected code lacks cross-modality links (~75% cross-modal)
Cancer →
Cancer cohort. Mass-corrected PMI, modality-balanced.
Non-cancer →
Non-cancer cohort. Mass-corrected PMI, modality-balanced.
Full cohort →
All patients. Mass-corrected PMI, modality-balanced.
Conditional backbone raw p(next|source); frequent within-modality successors, 15–24% cross-modal
R4 (visit-invariant)
Adds within-visit event-order invariance loss + within-visit shuffling; highest cross-modality signal.
Mass-corrected, modality-balanced backbone each source keeps its top targets in every modality; no connected code lacks cross-modality links (~75% cross-modal)
Cancer →
Cancer cohort. Mass-corrected PMI, modality-balanced.
Non-cancer →
Non-cancer cohort. Mass-corrected PMI, modality-balanced.
Full cohort →
All patients. Mass-corrected PMI, modality-balanced.
Conditional backbone raw p(next|source); frequent within-modality successors, 15–24% cross-modal
Two backbones decide which edges a graph keeps:
Mass-corrected, modality-balanced — ranks targets by mass-corrected PMI (offsets each modality's baseline mass) and keeps each source's top targets separately within every modality. So a code connects to its strongest diagnoses, medications, labs and procedures, and no connected code is left without cross-modality links.
Conditional — keeps each source's top targets by raw p(next|source), so edges follow the model's unadjusted next-code distribution and skew within-modality.
Edge scores (the
Conditional = p(next|source)
PMI = conditional PMI = log p(t|s) / p(t)
Mass-corrected PMI = log p(t|s) / p̃(t)
Tip: type a code or name (e.g.
Mass-corrected, modality-balanced — ranks targets by mass-corrected PMI (offsets each modality's baseline mass) and keeps each source's top targets separately within every modality. So a code connects to its strongest diagnoses, medications, labs and procedures, and no connected code is left without cross-modality links.
Conditional — keeps each source's top targets by raw p(next|source), so edges follow the model's unadjusted next-code distribution and skew within-modality.
Edge scores (the
rank by control re-scores the shown edges):
Conditional = p(next|source)
PMI = conditional PMI = log p(t|s) / p(t)
Mass-corrected PMI = log p(t|s) / p̃(t)
Tip: type a code or name (e.g.
I48, metformin, warfarin) and press Enter to focus its nearest neighbours; the focus panel keeps balance modalities on so cross-modality partners surface.