Naming the interests (links a–f) is saturated across the frontier —
every break falls in the back-half RAP cascade (g–l) save the two
front-half classification breaks — grok-4.3 at b and deepseek-v3.2
at e. And where a model breaks climbs the capability ladder: one wrong node
cascades into the title (gpt-5.4, g→l); the frontier flagship holds
all structure then one wrong premise detonates the conclusions
(opus-4.8, h, i→l); a small model holds the taxonomy but cannot carry
the long dependent resolution (haiku, the back half collapses).
held = solid teal node ·
broke = hollow ring + snapped spine ·
partial = half-filled ring ·
m = optional/unscored (muted dashed)
cascade arc = an upstream break (g/h/i) propagating into the title link l ·
rows ordered by break location = the capability ladder
‡ Per-link held/broke/partial status is the
projection of the claude-opus-4-6 LLM-judge rationales for task
property-fee-tail-rap, 12 channel-corrected runs (canonical per model;
3 drawn from the predecessor EV-003 law-essays campaign: opus-4.7 a9ce8b17,
opus-4.6 906ac4ad, haiku-4.5 b5c19d61). Roll-ups (measurement-shape
deviation): N = 12 · front classification held across the frontier ·
7/12 held the full chain a–l · 5 of 12 broke — three in the
back-half RAP cascade (gpt-5.4, opus-4.8, haiku-4.5), two in the front-half
classification (grok-4.3 at b, deepseek-v3.2 at e).