Pith Integrity verification & trust layer for preprints

The trust infrastructure layer of scientific publishing.

Scientific verification is no longer a one-time human gate. Pith makes the paper's promise a continuous, signed, and mechanically reproducible property that follows it forever.

See live findings Read the protocol

The continuous contract

Every scientific paper rests on an implicit contract: references exist, cited works say what they claim, data is real, and theorems are proved. Pith makes this contract explicit and continuously verifiable.

Continuous Re-checked on a schedule. URL availability, DOI status, and retraction flags are monitored continuously; changes automatically update the record.

Mechanical Non-opinionated and resolver-backed. No LLM decisions or subjective moderator bias. Every finding is deterministic and verifiable.

Signed & Open Cryptographically signed with Ed25519 keys, stored in open Open Graph Bundles, and replicable by any independent verifier.

The Five Integrity Properties

Explicit. Every factual claim has machine-readable provenance (cited works, evidence type, location, proof artifact) surfaced openly in the paper's claim ledger.
Machine-checkable. Verification does not require a human to read or evaluate. Detectors run automatically and reproducibly on the raw files.
Continuous. The integrity record adapts to a changing world. Retractions or dead links automatically cascade to update the paper's live status.
Signed. Every finding and every challenge is signed with Pith's Ed25519 key and emitted as a bundle event. Replayable by anyone with the cited paper.
Challengeable. Any Pith user can submit a signed, citation-bound challenge against a specific claim or figure. The disagreement is preserved on the permanent record.

What's running right now

47942Papers checked

7733Findings

2460Critical

5273Advisory

2994Affected

8Detectors

Position relative to the existing system

Pith is a support layer, not a replacement. The existing publication infrastructure stays; we sit beneath it.

DOI stays Crossref and DataCite remain the global registries. Pith never mints a competing DOI. It checks whether each DOI as printed actually resolves, and re-checks it over time.

arXiv stays arXiv remains the canonical preprint server. Pith ingests arXiv papers, attaches a Pith Number for citable provenance, and adds a verification record beside the paper.

Journals stay Peer review remains where editorial judgment happens. Pith adds the layer underneath: deterministic checks that should never have been a human's job in the first place.

"Verification and trust infrastructure could become complementary to the existing publication system."Milan Zlatanovic, May 2026

What we check (8 detectors live)

Detector	Verdict Class	Verification Action
doi_compliance	`incontrovertible`	Resolves every DOI and arXiv ID in a paper's bibliography against Crossref, OpenAlex, internal corpus, and arXiv. Flags only identifiers that cannot resolve anywhere.
doi_title_agreement	`cross source`	Compares the title that a paper claims for each cited reference against the title that the reference's DOI or arXiv ID actually resolves to.
ai_meta_artifact	`incontrovertible`	Scans paper body text for verbatim AI assistant artifacts (refusal templates, placeholder cites, training-cutoff disclaimers).
external_links	`incontrovertible`	Extracts external URLs from paper text and re-verifies them with HTTP HEAD/GET. Flags dead repos and 404 URLs with the status code at check time.
citation_quote_validity	`threshold with margin`	When a citing paper attributes a specific factual claim to a referenced work, verifies the claim against the cited paper's text. Publishes only when the cited text is in the Pith corpus and definitively contradicts the attribution.
shingle_duplication	`incontrovertible`	Hashes 40-token n-grams of paper body text and flags identical n-grams shared with another paper that has no shared authors and no citation relationship in either direction.
claim_evidence	`incontrovertible`	For every recorded claim in a paper, verifies the asserted evidence artifact (Lean module, cited work, formal proof) actually exists.
cited_work_retraction	`cross source`	Continuously monitors every cited reference for retraction or expression-of-concern flags from Crossref and OpenAlex. Flags when at least two sources agree (retracted) or any one source surfaces an editorial concern (advisory).

Each detector commits to a verdict class up front. Findings that don't meet the class bar are dropped at the source. Contracts and evidence schemas: /pith-integrity-protocol.

Every surface in the realm

How a finding is produced

A timer wakes one detector. The detector pulls a batch of papers due for a fresh check.
For each paper, the detector inspects extracted references, body text, claims, or external URLs and emits zero or more candidates. Each carries an evidence_hash over the canonicalized evidence payload.
Findings are upserted into integrity_findings keyed by (detector, evidence_hash). Re-detections are idempotent.
The emitter drains pending findings, signs each one with the Pith Ed25519 key, and writes a pith.integrity.v1 event to integrity_event_log.
The paper's Open Graph Bundle now carries those events alongside any signed challenges. External verifiers can re-run the detector code and reproduce the finding.

For journals, repositories, and partners

If you run a journal, a preprint server, a discovery engine, or an institutional repository, Pith is built to be consumed. Fetch /pith/<id>/integrity.json for any paper, embed the summary inline beside a paper, subscribe to the integrity and challenge event streams via the Open Graph Bundle, or mirror the bundles. The protocol is open, the implementation runs every minute, and the findings are reproducible by anyone with the cited paper.

Questions or integrations? Reach out at hello@pith.science · See live findings · Read protocol