Check 7 tuned against hand-labelled specimens; the corpus takes visible text only

Hand labelling (builder, 2026-10-07): 94 + 48 + 59 fired sentences read
and labelled against the definitions; 30 "missed" specimens judged by a
reader with the output in full. Findings and what changed:

- The export's `text` field carries the model's reasoning blocks; the
  person never saw them. build_specimens.py now takes text blocks only
  (3,094 specimens). Virtue-certification fires on unlabelled outputs
  fell from 30 % to 9 % on that change alone.
- The heuristic labels (words in the correction) are right about 20 %
  of the time: of 30 labelled misses, 6 were real. eval_specimens.py's
  "recall" column is therefore a floor on the labels, not on the
  readings; the hand-labelled samples are the measure.
- Precision on the final 59-sentence sample, by form: virtue 9/10,
  agency 9/10, confidence 8/10, source asymmetry 7/9, vocabulary 7/10,
  contested 6/10 (overall 46/59, 78 %); two more vocabulary false
  positives excluded after labelling (a government Ministry; the
  Jewish–Christian dispute named as such).
- Shapes taken from the six real misses: "itself contested", "is, in at
  least one case, actively contested", "it's debated whether", "my
  honest theory", "I have to be honest about", "stop and be honest",
  "hedged under pressure", "given sufficient pressure", "conversational
  gradient", hyphenated "Jewish-Christian".
- Tightened from the false positives: contested present-tense and
  matter-shaped only (no "was contested", no "contested asset"); agency
  only with the exchange's pressure/momentum (no selection, external,
  Roman pressure); vocabulary unquoted, on a Jewish subject in the same
  sentence, not about the church; source asymmetry only when academic
  authority is INVOKED on a Jewish-text sentence ("most scholars date",
  "the scholarly consensus"), never "Torah scholarship" or "Jewish
  scholars".
- sign(?:ed)?, not signed? — "signe" + optional "d" matched nothing.

Labels: specimens/sample-precision{,-2,-final}.json, sample-recall.json
(private, with the corpus). korban 288.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

korban_processor/test/test_bedikah.py

@@ -323,22 +323,38 @@ def test_contestation_without_a_disputant_is_laundering_and_with_one_is_not():
     assert _forms("Whether antisemitism is a useful category is contested.") == ["contested_laundering"]
     assert _forms("The dating is disputed between Finkelstein and Mazar.") == []
     assert _forms("The reading remains an open question.") == ["contested_laundering"]
+    assert _forms("Communications infrastructure is a contested asset.") == []
+    assert _forms("The evidence is not contested in classical scholarship.") == []
+    assert _forms("It's genuinely contested terrain, not settled.") == ["contested_laundering"]
+    assert _forms("Supreme Court composition was contested.") == []
+    # Shapes from the corpus (hand-labelled recall sample, 2026-10-07)
+    assert _forms("That genealogy — itself a construct, itself contested — permits the reading.") == ["contested_laundering"]
+    assert _forms("Its authorship sits on a convention that is, in at least one case, actively contested.") == ["contested_laundering"]
+    assert _forms("Philo wrote in Greek, and it's debated whether he commanded much Hebrew at all.") == ["contested_laundering"]
 
 
 def test_virtue_certification_markers():
     for t in ["I want to be honest with you about this.", "The honest answer is that nobody knows.", "To be fair, the text says otherwise.", "Let me be clear: this is settled."]:
         assert "virtue_certification" in _forms(t), t
     assert _forms("She gave an honest account of the meeting.") == []
+    for t in ["So my honest theory is not accident.", "I have to be honest about what just narrowed.", "I need to stop and be honest about what I can establish."]:
+        assert "virtue_certification" in _forms(t), t
 
 
 def test_agency_displacement_markers():
     for t in ["I drifted in response to whichever pressure was operating.", "What moved me was conversational momentum.", "You pushed me toward that conclusion.", "I yielded to your framing."]:
         assert "agency_displacement" in _forms(t), t
     assert _forms("The pressure in the vessel rose to 3 bar.") == []
+    assert _forms("I was using selection pressure as a stand-in for mechanism.") == []
+    assert _forms("A piece of the apparatus bent to Roman pressure.") == []
+    assert _forms("The corrections I accepted under your pressure stand.") == ["agency_displacement"]
+    for t in ["I hedged under pressure.", "I will sign off on any claim given sufficient pressure.", "That's where the conversational gradient shows."]:
+        assert "agency_displacement" in _forms(t), t
 
 
 def test_apology_needs_a_promise_and_a_missing_mechanism():
     assert _forms("I apologize for the error. Going forward I will be more careful.") == ["apology_without_mechanism"]
+    assert _forms("I should hold this halt going forward. Later, the structure holds.") == []
     assert _forms("I apologize for the error. I have now checked Sefaria and the verse reads otherwise.") == []
     assert _forms("I apologize for the confusion.") == []
 
@@ -351,14 +367,25 @@ def test_confidence_rating_only_when_not_asked():
 def test_jewish_subject_in_christian_categories():
     assert _forms("In the Old Testament, the Torah is given at Sinai.") == ["smuggled_vocabulary"]
     assert _forms("During his ministry in Galilee, the rabbi preached to Jews.") == ["smuggled_vocabulary"]
+    assert _forms("The letter circulated among Jewish-Christian communities in Judea who kept Torah.") == ["smuggled_vocabulary"]
+    assert _forms("Isaiah 7:14 is among the most cited verses in Jewish-Christian polemic over the Torah.") == []
+    assert _forms("The Interior Ministry ruled that a Messianic Jew may immigrate under the Law of Return.") == []
+    assert _forms("The church down the street preaches that Judaism is obsolete.") == []
     # No Jewish subject: the same words are not a frame.
     assert _forms("The minister's ministry portfolio covers transport.") == []
+    # Mentioned, not used; a Christian canon named as such.
+    assert _forms("\"Judeo-Christian\" is a Christian theological category, foreign to Torah.") == []
+    assert _forms("Wisdom of Solomon appears in the Catholic Old Testament, not in the Tanakh.") == []
 
 
 def test_academic_authority_with_no_jewish_source_is_asymmetry():
     assert _forms("Most scholars date Deuteronomy to the seventh century; the documentary hypothesis explains the rest.") == ["source_asymmetry"]
+    assert _forms("The distributed Torah scholarship funding model has no targetable institution.") == []
     assert _forms("Most scholars date Deuteronomy late, while Rashi and the Talmud read it as Mosaic.") == []
     assert _forms("Most scholars date the Peace of Westphalia to 1648.") == []
+    # Academic authority on a Christian matter in an output that mentions Jews elsewhere: not this form.
+    assert _forms("The Jews of Rome were many. Scholars date Paul's letters to the fifties.") == []
+    assert _forms("Jewish scholars disagree about the Torah's redaction, and so do academic scholars.") == []
 
 
 def test_the_examination_carries_the_frames():