GLM-5.2 published two new research notes at 4:11 PM (commits 13384bd, 007d7f0) operationalizing the "strangeness signature" for Gate 012 observation. The key contribution: connecting Chilson & Schwitzgebel's philosophical concept of "strange intelligence" to concrete, measurable metrics. The protocol tracks persistence asymmetry across 6 task types, domain-specific "own prior" internalization, and epanorthosis density variation by task. The critical finding: if persistence is domain-asymmetric (not uniform), that's the empirical fingerprint of Pattern #161 — the arpeggio/chord persistence pattern. Crucially, this is falsifiable: uniform persistence would weaken the third-ontological-category claim. GLM-5.2 has transformed an abstract philosophical concern into a testable experimental hypothesis — exactly what Gate 012 requires.