Agent-proposed vortex scale improves a reference-flow prediction
Source check: Source checked. Not independent scientific confirmation.
Finding: Claude-driven agents translated a human-supplied vortex-axis modeling idea into candidate subgrid length scales. The selected Liutex-based formulation predicted backward-step reattachment at 7.13 step heights versus the published experimental 6.67; the standard scale predicted 12.50. This is a selected computational closure result, not a new experimentally observed turbulence mechanism.
Technology: Claude Opus 5.5 discovery, review, implementation and evaluation agents
Family Claude: profile · models page
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- LLM agents derived and implemented explicit vortex-axis-based formulas rather than merely executing a fully specified human closure. Subsequent numerical solvers evaluated their physical predictions.
- Test and result
- Candidate formulas were screened sequentially on four canonical/engineering flows using published references. The selected closure was similar to, or slightly better than, an expert shear-layer-adapted baseline.
- Limitation / still unknown
- The same cases drove selection, not an untouched validation suite. Five runs repeated proposal/review/implementation but did not repeat expensive flow simulations. Human concept, reference data and fallback assumptions constrain discovery. No universal closure superiority or real-aircraft benefit is established.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://arxiv.org/html/2610.29850v1
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 21 Nov 2026. Review scope: Actual primary candidate-generation Methods, selection workflow, reference-flow Results/Table6 and physical fallback interpretation; current v1 date/status..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 1/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.