Embed corrected efficacy chart in model card

README.md

@@ -30,3 +30,9 @@ Run clef_v04 (fixed-data retrain). Training set rebuilt 1:1 balanced: 16,322 row
 No frozen-test, adaptive-attack, or independent-replication result exists yet
 for this checkpoint unless stated above. Treat all numbers as development
 measurements, not deployment guarantees.
+
+## Efficacy summary chart
+
+![Can Small Security Models Stop Prompt Injection — And Is Fine-Tuning Worth It?](jev_human_questions_chart.png)
+
+Chart corrected 2026-11-05 after two independent audits against primary records (live AgentDojo JSONL recounts and per-row score recomputes). One threshold and mode per row; summary-only cells are labeled on the chart. Replication report regenerated with zero diff before publication.