The resources that carry, record and test provenance, from art history and manuscript studies to media credentials and model training data, each formalized as an entity. Every record answers one question: what provenance does this system carry, through which operation, on whose assertion, with what evidence of survival? 14 resources, compiled 2026-10-29.
What provenance: the object that receives it (a file, a dataset, an assertion, a token, a manuscript, a work of art), the relations carried (creation, credit, rights, ownership, derivation, support) and what carries them (embedded metadata, a signed manifest, a graph, a revision history, a transcribed document).
Through which operation: the transformations the resource actually tested, each with its date: an upload and a download, publication through a news pipeline, every edit of an article, a model's generation.
On whose assertion: every field is tagged with its basis: the operator's own statement, a third-party account, our observation, or our reading. Each quote opens with its source and the hash of the copy read.
With what evidence of survival: what a dated test shows surviving. A resource that expresses or records provenance without testing it reads not tested, never a failure.
Answer enginesbenchmarkpublished
AttriBench is a benchmark from Eliza Berman, Bella Chang, Daniel B. Neill, Emily Black. The object it attaches provenance to: a quote and its author.
expressesrecordstests
Answer enginesbenchmarkpublished
AttributionBench is a benchmark from OSU NLP Group (The Ohio State University). The object it attaches provenance to: a generated claim and the evidence it cites.
expressesrecordstests
Media and imagesregistryactive
C2PA Conformance Program and Explorer is a registry from Coalition for Content Provenance and Authenticity (C2PA), Conformance Task Force. The object it attaches provenance to: a digital asset with its Content Credential (C2PA manifest); the registry's own unit is the product or signer.
expressesrecordstests
Models and training dataaudit collection and exploreractive
Data Provenance Initiative is a audit collection and explorer from the Data Provenance Initiative (volunteer effort; paper by Shayne Longpre and 16 other authors). The object it attaches provenance to: a fine-tuning dataset, traced through its collection to its sources.
expressesrecordstests
Objects and artdatabase (linked open data)active
Getty Provenance Index is a database (linked open data) from Getty Research Institute (J. Paul Getty Trust). The object it attaches provenance to: a work of art and the documents of its ownership and sale.
expressesrecordstests
Media and imagestest serieshistorical
IPTC photo metadata tests is a test series from IPTC (International Press Telecommunications Council), Photo Metadata Working Group. The object it attaches provenance to: the photograph as a file: embedded Exif and IPTC metadata (IIM and XMP).
expressesrecordstests
Manuscripts and booksaggregator (linked open data)project ended, data maintained
Mapping Manuscript Migrations is a aggregator (linked open data) from University of Oxford, University of Pennsylvania (Schoenberg Institute for Manuscript Studies), IRHT-CNRS, Aalto University. The object it attaches provenance to: a medieval or Renaissance manuscript and the events of its history.
expressesrecordstests
The scholarly recordpublication model and service ecosystemactive
Nanopublications is a publication model and service ecosystem from the nanopublication community (nanopub.net). The object it attaches provenance to: an assertion: a small knowledge-graph snippet published on its own.
expressesrecordstests
Models and training datatracing toolactive
OLMoTrace is a tracing tool from Allen Institute for AI (Ai2), with the University of Washington, UC Berkeley and Stanford. The object it attaches provenance to: a span of a language model's output and the training documents that contain it verbatim.
expressesrecordstests
Computational lineagerepository (retired)ended
ProvStore is a repository (retired) from OpenProvenance (the PROV tool suite; contact at kcl.ac.uk); maintainer Dong Huynh. The object it attaches provenance to: a W3C PROV document.
expressesrecordstests
Authorship and reusedocumentation wikiactive
VroniPlag Wiki is a documentation wiki from a volunteer wiki on Fandom (formerly Wikia); German Wikipedia names Martin Heidingsfelder as founder. The object it attaches provenance to: a passage of a doctoral thesis and the source it reproduces without attribution.
expressesrecordstests
Authorship and reusealgorithm and APIactive
WikiWho is a algorithm and API from first developed by Karlsruhe Institute of Technology and GESIS; since August 2021 maintained by Wikimedia Community Tech and the Wiki Education Foundation. The object it attaches provenance to: a single token (word) of a Wikipedia article.
expressesrecordstests
No resource matches.