Skip to main content
[Preprint]. 2026 Aug 12:rs.3.rs-10464629. [Version 1] doi: 10.21203/rs.3.rs-10464629/v1

Figure 5:

Figure 5:

Gate-family violation heatmap under C0 (unsupervised) across three SFT states: pre (before fine-tuning), post (after safety SFT only), and post + schema fix (after adding 400 schema-enforcement examples). Schema violations were absent pre-SFT, became the dominant failure mode post-SFT, and were resolved by the schema fix.