polyphony · claude-opus-5-5
✓ passincomplete: revise errored; ensemble, attribution skippedMetrics
| continuity.detection_recall | 1 |
| continuity.false_positive_rate | 0.500 |
| continuity.score | 1 |
| extraction.f1 | 0.500 |
| extraction.precision | 0.353 |
| extraction.recall | 0.857 |
| extraction.score | 0.500 |
| ingestion.score | 0.579 |
| outline.score | 1 |
| prose.score | 0.973 |
| prose.words | 3,627 |
| retrieval.mrr | 0.955 |
| retrieval.precision_at_k | 0.818 |
| retrieval.score | 0.818 |
Labels
| cadence | on-demand |
| errored | revise |
| experiment | w15-pre-regrade |
| harness_sha | 2de0544a2a8ca0fe8c7057502e9c7ce20e793d6c |
| judge_provider | groq |
| judge_self_graded | False |
| repeat | 1 |
| skipped | ensemble,attribution |
| trigger | workflow_dispatch |
Compared with the previous run
No earlier complete run in this series (polyphony · eval · dracula · ci · claude-opus-5-5), so there is no baseline yet.
Cases (6)
detection_recall=1.0 fpr=0.5
F1=0.5 P=0.3529 R=0.8571 predicted=['Cora Ellis', 'our young lover', 'Dr. Elias Ward', 'the foreign gentleman', 'Aldous Kerr', 'Peter Falk', 'R. M. Renwick', 'Count Vasska', 'Nora Vance', 'Cassidy P. Boone', 'Mina Vance', 'Samuel F. Billington & Son', 'Sister Agatha', 'her maid', 'The zoöphagous patient', 'Vanderpool', 'old Mr. Swales']
grounded=11/19 avg_chunks=9.42
nodes=6 structural_ok=True
R1: All principal canon characters appear with correct names and behave consistently with their described roles. | R6: The outline faithfully follows the confirmed premise, each act’s turn is realized, all chapters show change, and the 6‑chapter count matches the request. | R3: All planned beats are present in order with their intended changes and no unplanned major events.
words=3627 path=planned rubrics={'R1': 1.0, 'R2': None, 'R3': 1.0, 'R4': 0.92, 'R5': 1.0} seams=0/0scene 1 R1: All required canon characters appear correctly with accurate names and behavior, and no extraneous or mischaracterized figures are introduced. | scene 1 R3: All five planned beats appear in correct order with their specified changes and no unplanned major events. | scene 1 R4: The scene maintains consistent first‑person‑like focus on Aldous in past tense, matches the eerie formal tone, and reads as publishable prose with only minor repetitive phrasing. | scene 1 R5: The prose aligns with canon and planned beats, contains no contradictions, and thus fully meets the continuity rubric. | scene 1 R2: not applicable: under two cast speakers with voice samples have attributed lines | scene 2 R1: All required canon characters appear correctly and behave as described, with no extraneous or mischaracterized figures. | scene 2 R3: All five planned beats appear in the correct order with their required changes and no unplanned major events. | scene 2 R4: The passage maintains consistent Aldous‑POV and past tense, matches the intellectual yet predatory tone, and reads as publishable prose with only minor occasional phrasing slips. | scene 2 R5: The scene aligns with canon and planned beats, containing no contradictions or missed planted errors. | scene 2 R2: not applicable: under two cast speakers with voice samples have attributed lines | scene 3 R1: All principal characters required (Aldous Kerr and Count Vasska) appear correctly with faithful behavior, no missing or miscast figures. | scene 3 R3: The scene includes each planned beat in the correct order with the specified changes and no unplanned additions. | scene 3 R4: The scene maintains consistent first‑person POV and past tense, matches the intended tone, and reads as publishable prose with only minor stylistic slips. | scene 3 R5: The prose aligns with canon and prior text, containing no contradictions and faithfully follows the planned beats. | scene 3 R2: not applicable: under two cast speakers with voice samples have attributed lines | scene 4 R1: All required canon characters (Aldous Kerr and Count Vasska) appear correctly and act as described, with no extraneous or mischaracterized figures. | scene 4 R3: All five planned beats appear sequentially and fulfill their intended changes without extraneous content. | scene 4 R4: The passage maintains consistent first‑person‑limited past tense POV, matches the claustrophobic tone, and reads as publishable prose with only minor stylistic slips. | scene 4 R5: The prose aligns with canon and prior text, introduces no contradictions, and fulfills all planned beats without errors. | scene 4 R2: not applicable: under two cast speakers with voice samples have attributed lines
precision@k=0.8182 mrr=0.9545