944 gate-passed questions · 25 cells · 7 occupied · go to this type's cells
Bins were cut on this corpus at 240×135@16 (CLEAR_PX 25, motion floor 19 px). Margin bins: a near-edge band [1, 1.25) above the gate boundary, then corpus quantiles; salience bins: a visibility floor, then quantiles.
yes: peak_px_max / CLEAR_PX; no: one bin · measurable on 538 questions
| margin bin | numeric range |
|---|---|
| near-edge [1,1.25) | [−∞, 1.25) |
| bin0 | [1.25, 25.52) |
| bin1 | [25.52, 52.04) |
| bin2 | [52.04, ∞) |
| state:absent | (state / single bin) |
Ask: is this the same quantity the gate certifies on for this type? Are the states (e.g. 'same', 'gone') the right one-bin exceptions? Are the edges where a change in difficulty actually happens?
min queried-object peak area (frac of frame) · floor 0.0007716
| salience bin | numeric range |
|---|---|
| <floor (0.0008) | [−∞, 0.0007716) |
| bin0 | [0.0007716, 0.01543) |
| bin1 | [0.01543, 0.02886) |
| bin2 | [0.02886, 0.04747) |
| bin3 | [0.04747, ∞) |
Ask: does this axis capture what makes a question of this type perceptually hard, independently of the answer? Is the floor the right anti-gaming boundary?
answer (branch 1) · labels: no, yes
Compatible margin bins per context label (cells outside these are definitionally impossible):
| answer | n |
|---|---|
| yes | 538 |
| no | 406 |