Compare MTP steps and prefill batches; retain qualified MTP 2 baseline

2026-09-18 23:22:45 +08:00
parent 918b78a3b8
commit c8bfe6b9ab
45 changed files with 3785 additions and 3 deletions
@@ -0,0 +1,8 @@
+{"name": "invoice-extraction", "passed": true, "elapsed_s": 1.2744407369973487}
+{"name": "deduplicate", "passed": true, "elapsed_s": 1.7563880499947118}
+{"name": "schedule", "passed": true, "elapsed_s": 2.0583331399975577}
+{"name": "chinese-policy", "passed": true, "elapsed_s": 1.335008790003485}
+{"name": "merge-intervals", "passed": true, "validation_stdout": "passed\n", "validation_stderr": ""}
+{"name": "binary-search", "passed": true, "validation_stdout": "passed\n", "validation_stderr": ""}
+{"name": "tool-arguments", "passed": true}
+QUALITY_COMPLETE