File tree Expand file tree Collapse file tree
reproduction-results/results_v2 Expand file tree Collapse file tree Original file line number Diff line number Diff line change @@ -99,3 +99,18 @@ yields a degenerate output. Two distinct failure modes therefore coexist:
9999
100100Legend: ✅ exact syntactic match; ⚠️ prompt exceeded 4096 tokens (FIM truncation,
101101output degenerate). Predicates are normalised (whitespace-insensitive).
102+
103+
104+ ## Summary distribution over the 28 contracts
105+
106+ * Equivalent to GT* = the tool recovered a guard equivalent to (or
107+ stronger than) the patch. * Not equivalent to GT* includes the case where the
108+ patch guard is differente from the one deployed by the tool.
109+ * Truncated / error* = every hole is a degenerate FIM output.
110+
111+ | Result (per contract) | Count | % |
112+ | ---| ---| ---|
113+ | Equivalent to GT | 6/28 | 21.4% |
114+ | Not equivalent to GT | 11/28 | 39.3% |
115+ | Truncated / error | 11/28 | 39.3% |
116+ | ** Total** | ** 28/28** | ** 100%** |
You can’t perform that action at this time.
0 commit comments