{"paper":{"title":"Accelerating point defect simulations using data-driven and machine learning approaches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Data-driven machine learning models trained on DFT calculations accelerate point defect simulations while retaining quantum-mechanical accuracy.","cross_cats":["physics.chem-ph","physics.comp-ph"],"primary_cat":"cond-mat.mtrl-sci","authors_text":"Arun Mannodi-Kanakkithodi, Menglin Huang, Prashun Gorai, Se\\'an R. Kavanagh","submitted_at":"2026-04-22T20:29:41Z","abstract_excerpt":"Point defects in solid-state materials are now routinely simulated using large supercell structures, requiring efficient quantum mechanical solutions. Data-driven and machine learning (ML) models trained on computational data can enable rapid defect property predictions and high-throughput screening. In this article, we provide an overview of prominent efforts to accelerate defect simulations using these approaches. We begin by discussing the motivations for data-driven techniques in defect modeling, and describe efforts over the past decade to use descriptor-based models for rapid screening o"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Surrogate models and interatomic potentials trained on density functional theory data lead to predictions with quantum-mechanical accuracies at a fraction of the cost, including for phonon modes and vibrational free energies at finite temperatures.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the machine learning models trained on limited DFT data can generalize accurately to new defect configurations and materials not seen in training.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Machine learning models trained on quantum mechanical data can predict defect properties in solids with high accuracy but at much lower computational cost than traditional methods.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Data-driven machine learning models trained on DFT calculations accelerate point defect simulations while retaining quantum-mechanical accuracy.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8b5d49e32791dc2e19d4e68324fe9e08747b989885da003b2f683df4190c1a9d"},"source":{"id":"2604.21069","kind":"arxiv","version":1},"verdict":{"id":"cf8140b9-a412-4540-92c9-3d2eff6b7800","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T23:26:20.587538Z","strongest_claim":"Surrogate models and interatomic potentials trained on density functional theory data lead to predictions with quantum-mechanical accuracies at a fraction of the cost, including for phonon modes and vibrational free energies at finite temperatures.","one_line_summary":"Machine learning models trained on quantum mechanical data can predict defect properties in solids with high accuracy but at much lower computational cost than traditional methods.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the machine learning models trained on limited DFT data can generalize accurately to new defect configurations and materials not seen in training.","pith_extraction_headline":"Data-driven machine learning models trained on DFT calculations accelerate point defect simulations while retaining quantum-mechanical accuracy."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.21069/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T13:38:05.276058Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-20T01:20:45.505796Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"68ca2cd1eceb18c6d16d3a7a66bdf56d1e1144c9b677031d71a172265b4f8036"},"references":{"count":143,"sample":[{"doi":"","year":2022,"title":"L. Vines, E. Monakhov, and A. Kuznetsov, “Defects in semiconductors,” Journal of Applied Physics, vol. 132, p. 150401, 10 2022","work_id":"ecc12e56-12b3-41b2-83ea-90ad182f2c8e","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Perspective on defect control in semiconductors for photovoltaics,","work_id":"bf74c709-1f28-460b-8147-e644c51916c8","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"The defect challenge of wide-bandgap semiconductors for photovoltaics and beyond,","work_id":"65f9dd48-f3d8-44dc-9569-f347f2b669bc","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Pycdt: A python toolkit for modeling point defects in semiconductors and insulators,","work_id":"e2a321a5-c349-4cd1-905a-188733fa4076","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Point defect engineering in thin-film solar cells,","work_id":"b47513a1-57da-4cba-9dba-ebd9d9db2452","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":143,"snapshot_sha256":"db830fd9e1100f067affcb1dcfed8d2a83a7822e6effa7fca567737ef437f2b0","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}