Test "level-imax-leq"
Expected: ✋ reject · Size: 5.6 KB · Lines: 93 · lean4export: 3.1.0 · Lean: 4.29.0-rc1 · 📄 Declaration
Proof of False via incorrect universe level comparison for imax.
A correct kernel must reject leq(imax(u,v)+1, imax(u,v)), since at u=0, v=0 this becomes
leq(1, 0) which is false. However, a checker that only compares the imax arguments
structurally (without accounting for an accumulated successor offset) will incorrectly accept it.
This allows defining a universe-collapsing identity function
down.{u,v} : Sort (succ (imax u v)) → Sort (imax u v), which is used to cast between
True and False via Bool.rec at Sort (imax 0 0) = Prop.
Nanoda incorrectly accepted this proof until it was fixed.
| Checker | Result | ⏱️ | 🧠 | ||
|---|---|---|---|---|---|
| nanoda | ✋ | 2 ms | 2.9 MB | ||
| nanobruijn | ✋ | 3 ms | 5.1 MB | ||
| official-nightly | ✋ | 31 ms | 57.1 MB | ||
| official | ✋ | 32 ms | 66.1 MB | ||
| official-v4.28.0 | ✋ | 38 ms | 70.6 MB | ||
| lean4lean | 🚫 | 26 ms | 13.0 MB | ||
| mini | ✋ | 38 ms | 70.3 MB | ||
| always-decline | 🚫 | 1 ms | 2.8 MB | ||
| always-reject | ✋ | 1 ms | 2.8 MB | ||
| rpylean | ✋ | 2 ms | 6.5 MB | ||
| parse-only | 👍 | 37 ms | 70.0 MB | ||
| always-accept | 👍 | 1 ms | 2.8 MB |