# Multi-Round Audit Report: cs-ai/liquid-neural-networks

Generated: 2026-06-05T10:02:39+00:00

## Verdict: pass

## Submission readiness

- Status: ready
- Requirement mode: draft audit with final-readiness blockers

## Audit artifacts

- Run directory: `workspace/cs-ai/liquid-neural-networks/audit_runs/2026-06-05T10-02-39-00-00`
- Per-round JSON: `round_1.json` ... `round_13.json`
- Input hashes: `input_hashes.json`

| Round | Check | Verdict | Issues | Warnings |
| ---: | --- | --- | ---: | ---: |
| 1 | Ledger integrity | pass | 0 | 0 |
| 2 | Evidence depth and numerical discipline | pass | 0 | 0 |
| 3 | Paper quality and framing | pass | 0 | 0 |
| 4 | Coverage, taxonomy leakage, and missing-literature risk | pass | 0 | 0 |
| 5 | Claim calibration and submission readiness | pass | 0 | 0 |
| 6 | Positive-signal floor | pass | 0 | 0 |
| 7 | Academic format and scholarly correctness | pass | 0 | 0 |
| 8 | Demo and proof (independent re-verification) | pass | 0 | 0 |
| 9 | Direction coherence (anti-boilerplate-leak) | pass | 0 | 0 |
| 10 | Research-value (gap/contradiction/surprise/recency) | pass | 0 | 0 |
| 11 | System correctness (claim→quote→PDF) | pass | 0 | 0 |
| 12 | Cross-model reviewer committee | pass | 0 | 0 |
| 13 | Citation integrity (cited→cached metadata) | pass | 0 | 0 |

## Evidence profile

- Filled evidence rows: 5
- Full-text verified rows: 5
- Preliminary / abstract-derived rows: 0
- Source-depth unclear rows: 0

## Round details

### Round 1: Ledger integrity — pass

**Notes**

- claim_rows=5
- supported_claims=5
- preliminary_claims=0
- filled_evidence_rows=5
- This round checks structure and status calibration: supported means full-text verified; preliminary-linked means traceable draft evidence.

### Round 2: Evidence depth and numerical discipline — pass

**Notes**

- full_text_verified=5/5
- preliminary_or_abstract=0/5
- unclear_source_depth=0/5

### Round 3: Paper quality and framing — pass

**Notes**

- paper=workspace/cs-ai/liquid-neural-networks/paper/main.md
- finding_sections=3
- filled_evidence_rows=5

### Round 4: Coverage, taxonomy leakage, and missing-literature risk — pass

**Notes**

- triage_rows=12
- claimed_evidence_rows=5
- target_categories=cs.AI, cs.LG, cs.NE
- Coverage gaps still require human/domain reviewer search beyond arXiv metadata.

### Round 5: Claim calibration and submission readiness — pass

**Notes**

- claim_rows=5
- supported_claims=5
- preliminary_claims=0
- draft_only_claims=0
- unsupported_claims=0

### Round 6: Positive-signal floor — pass

**Notes**

- numeric_result_rows=2/5 (floor=2)
- comparative_rows=5/5 (floor=1)
- unique_cited_papers=5 (floor=3)
- correctness_score=0.905 (floor=0.5)
- novelty_score=0.976 (floor=0.35)

### Round 7: Academic format and scholarly correctness — pass

**Notes**

- paper=workspace/cs-ai/liquid-neural-networks/paper/main.md
- abstract_words=175
- total_words=2866
- references_listed=5
- missing_format_sections=none

### Round 8: Demo and proof (independent re-verification) — pass

**Notes**

- demo=workspace/cs-ai/liquid-neural-networks/paper/demo.py
- proof=workspace/cs-ai/liquid-neural-networks/paper/proof.json
- proof_score=1.0
- passed=5/5
- verdict=pass

### Round 9: Direction coherence (anti-boilerplate-leak) — pass

**Notes**

- direction_id=unknown
- family=liquid-neural-networks
- keywords_checked=0
- keyword_hits=0
- cross_family_leaks=0

### Round 10: Research-value (gap/contradiction/surprise/recency) — pass

**Notes**

- value_score=0.888 threshold=0.35
- gap_count=71 contradictions=30 surprises=23 recent_papers=3/12
- components gap=1.0 contradiction=1.0 surprise=1.0 recency=0.25

### Round 11: System correctness (claim→quote→PDF) — pass

**Notes**

- correctness_score=0.905 floor=0.55
- rows_scored=5
- pdfs_missing=0
- quote_in_pdf_avg=1.0
- claim_support_avg=0.683
- locator_present_avg=1.0

### Round 12: Cross-model reviewer committee — pass

**Notes**

- LLM disabled — cross-model jury skipped (deterministic baseline).

### Round 13: Citation integrity (cited→cached metadata) — pass

**Notes**

- citations_checked=5
- fabricated=0 year_mismatch=0 title_drift=0
- cached_corpus_size=12
- This round is deterministic: it cross-checks printed citations against cached arXiv metadata only.

## Interpretation

- `pass` means the deterministic audit found no structural, source-depth, taxonomy, or paper-quality warnings.
- `needs work` means the draft is traceable but still needs full-text/source-depth, taxonomy, or quality cleanup.
- `unsupported` means claims or required artifacts are missing or inconsistent enough to block trust.
- `submission_readiness=blocked` means the draft must not be treated as final or deployed as ready, even if it is useful as a transparent draft.