{
  "abstract": "While current AI-driven methods excel at deriving empirical models from individual experiments, a significant challenge remains in uncovering the common fundamental physics that underlie these models -- a task at which human physicists are adept. To bridge this gap, we introduce AI-Newton, a novel framework for concept-driven scientific discovery. Our system autonomously derives general physical laws directly from raw, multi-experiment data, operating without supervision or prior physical knowledge. Its core innovations are twofold: (1) proposing interpretable physical concepts to construct laws, and (2) progressively generalizing these laws to broader domains. Applied to a large, noisy dataset of mechanics experiments, AI-Newton successfully rediscovers foundational and universal laws, such as Newton's second law, the conservation of energy, and the universal gravitation. This work represents a significant advance toward autonomous, human-like scientific discovery.",
  "abstract_shown": true,
  "affiliations": [
    "Peking University"
  ],
  "article_type": "article",
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "You-Le Fang",
      "orcid": "0009-0008-1661-2243"
    },
    {
      "affs": [
        0
      ],
      "name": "Dong-Shan Jian",
      "orcid": "0000-0002-7131-3181"
    },
    {
      "affs": [
        0
      ],
      "name": "Xiang Li",
      "orcid": "0009-0009-4946-1469"
    },
    {
      "affs": [
        0
      ],
      "name": "Yan-Qing Ma",
      "orcid": "0000-0001-5795-3791"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/2wtznxc1d5zww.bib",
  "cited_by": 8,
  "doi": "10.1103/dvvc-4rsr",
  "fulltext_url": "http://doi.org/10.1103/dvvc-4rsr",
  "id": "2wtznxc1d5zww",
  "issns": [
    "2643-1564"
  ],
  "landing_url": "https://link.aps.org/doi/10.1103/dvvc-4rsr",
  "language": "en",
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "feed-journal",
      "physics/general"
    ],
    [
      "lcc",
      "QC1-999"
    ],
    [
      "openalex",
      "3109|Model Reduction and Neural Networks"
    ]
  ],
  "openalex": "W4417233420",
  "pdf_url": "http://doi.org/10.1103/dvvc-4rsr",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-09",
  "publisher": "American Physical Society (APS)",
  "ris": "https://research.draconitic.com/paper/2wtznxc1d5zww.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:17:55Z",
      "native_id": "10.1103/dvvc-4rsr",
      "source": "crossref",
      "url": "https://link.aps.org/doi/10.1103/dvvc-4rsr"
    },
    {
      "first_seen": "2026-11-04T13:21:02Z",
      "native_id": "W4417233420",
      "source": "openalex",
      "url": "https://doi.org/10.1103/dvvc-4rsr"
    },
    {
      "first_seen": "2026-11-04T14:21:02Z",
      "native_id": "2b3144a3795a42b28468dc0eb1f86639",
      "source": "doaj",
      "url": "https://doi.org/10.1103/dvvc-4rsr"
    },
    {
      "first_seen": "2026-11-05T12:20:13Z",
      "native_id": "10.1103/dvvc-4rsr",
      "source": "feed-aps",
      "url": "http://link.aps.org/doi/10.1103/dvvc-4rsr"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "openalex:3109|Model Reduction and Neural Networks",
      "subfield": "statistical-nonlinear"
    }
  ],
  "title": "AI-Newton: A concept-driven physical law discovery system without prior physical knowledge",
  "title_text": "AI-Newton: A concept-driven physical law discovery system without prior physical knowledge",
  "type": "peer-reviewed",
  "updated_at": "2026-11-06T14:20:28Z",
  "url": "https://research.draconitic.com/paper/2wtznxc1d5zww",
  "venue": "Physical Review Research",
  "versions": []
}