# Fake Reference Checker - Find Hallucinated & Retracted Papers (`that_red_bird/fake-reference-checker`) Actor

Paste a bibliography, BibTeX or DOI list. Each reference is checked against Crossref, OpenAlex, DataCite and PubMed and marked found, mismatched or not found, with retracted, corrected and expression-of-concern flags. No API key. Books and very new papers can be missing.

- **URL**: https://apify.com/that_red_bird/fake-reference-checker.md
- **Developed by:** [mohamed alaya](https://apify.com/that_red_bird) (community)
- **Categories:** AI, Developer tools, Open source
- **Stats:** 2 total users, 1 monthly users, 100.0% runs succeeded, 0 bookmarks
- **User rating**: No ratings yet

## Pricing

Pay per event

This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events.

Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event

## What's an Apify Actor?

An Actor is a serverless cloud program that runs on the Apify platform. It has two run modes.
In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours,
and optionally produces a well-defined JSON output, datasets with results, or files in key-value store.
In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server.

Apify vocabulary and the platform model are defined once, in the agent quickstart at https://apify.com/agents.md.

## How to integrate an Actor?

If asked about integration, you help developers integrate Actors into their projects.
You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready.

Do not guess an integration path. Every one of them is in the agent quickstart at https://apify.com/agents.md: the Apify MCP server, Agent Skills with the Apify CLI, the JavaScript and Python clients, the REST API, and the account-free path for an agent with no human to sign in. It also carries the rule on stating cost before the first paid run.

For examples already wired to this Actor's own input schema, see the [API](#api) section below.

Each client library has reference documentation the quickstart does not restate: [JavaScript/TypeScript](https://docs.apify.com/api/client/js/docs.md) (`npm install apify-client`) and [Python](https://docs.apify.com/api/client/python/docs.md) (`pip install apify-client`).

# README

## Fake Reference Checker - Find Hallucinated & Retracted Papers

Paste a bibliography, a BibTeX file or a list of DOIs. Every reference comes back as **found**, **mismatched** (real paper, wrong authors, wrong title or wrong DOI) or **not found**, and papers that were **retracted** or have a correction or expression of concern are flagged in the same row.
Built for researchers, journal editors, universities and AI literature-review tools that need to catch fabricated citations and withdrawn work before they spread.

Checks run against Crossref (including Retraction Watch data), OpenAlex, DataCite and PubMed. No API key.

### Input examples

**1. Plain-text reference list**

```json
{ "references": "[1] Vaswani, A., et al. (2017). Attention is all you need. NeurIPS.\n[2] Smith, J., & Patel, R. (2022). Neural correlates of quantum memory in transformer hybrid models. Nature Machine Intelligence, 4(3), 211-224." }
```

**2. BibTeX**

```json
{ "references": "@article{key, author={Jumper, John and Evans, Richard}, title={Highly accurate protein structure prediction with AlphaFold}, journal={Nature}, year={2021}, doi={10.1038/s41586-021-03819-2}}" }
```

**3. DOIs only, show only the problems**

```json
{ "dois": ["10.1038/nature14539", "10.1016/S0140-6736(97)11096-0"], "onlyProblems": true }
```

Run with no input to see a demo: a real paper, a retracted paper, an invented paper, a real title with the wrong DOI, and a clean reference.

### Input guide

| Field | What it does |
|---|---|
| `references` | Text, numbered list, or BibTeX. Do not mix BibTeX and plain text in the same box. |
| `dois` | A list of DOIs or doi.org links. |
| `checkRetractions` | Add retracted / corrected / expression-of-concern flags (default on). |
| `useOpenAlex`, `usePubMed` | Extra indexes tried when Crossref has no match. |
| `contactEmail` | Optional; identifies you to Crossref and OpenAlex for faster service. |
| `onlyProblems` | Output only references that are not found, mismatched, retracted, or have a year difference. |

### How a reference is judged

1. If a DOI is present, it is looked up in Crossref, then DataCite (Zenodo, arXiv and data DOIs). A DOI that resolves to a different paper than the one cited is flagged `DOI_POINTS_TO_DIFFERENT_PAPER`.
2. Otherwise (or if the DOI does not settle it) the reference text is searched in Crossref, then OpenAlex, then PubMed.
3. The best candidate's title must appear in your reference text, its first author must match, and the year must be within one year. Authors or title that differ give `MISMATCHED`; no plausible candidate gives `NOT_FOUND`.

### Output

| Field | Meaning |
|---|---|
| `status` | `FOUND`, `MISMATCHED`, `NOT_FOUND` or `UNPARSEABLE` |
| `flags` | `RETRACTED`, `CORRECTED`, `EXPRESSION_OF_CONCERN`, `AUTHORS_DIFFER`, `YEAR_DIFFERS`, `TITLE_DIFFERS`, `DOI_NOT_FOUND`, `DOI_DIFFERS`, `DOI_POINTS_TO_DIFFERENT_PAPER` |
| `retracted`, `expressionOfConcern`, `corrected`, `retractionNotices` | Retraction status with the notice DOI and date |
| `matchedTitle`, `matchedAuthors`, `matchedYear`, `matchedDoi`, `matchedUrl`, `source` | The real paper found |
| `explanation` | Plain-English reason |

The `paper-report` row summarises the whole bibliography: counts, `suspectRate` and a `riskLevel`.

Sample row:

```json
{ "type": "reference", "index": 2, "status": "FOUND", "flags": ["RETRACTED", "CORRECTED"], "retracted": true,
  "matchedTitle": "RETRACTED: Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children",
  "matchedYear": 1998, "matchedDoi": "10.1016/s0140-6736(97)11096-0", "source": "crossref",
  "retractionNotices": [{ "type": "retraction", "doi": "10.1016/s0140-6736(10)60175-4", "date": "2010-02-06", "source": "retraction-watch" }] }
```

### Use it with

- **LLM literature-review tools:** pass the model's bibliography here and drop anything that is not `FOUND`.
- **Journal submission checks:** run the manuscript's reference list; send `MISMATCHED` and `NOT_FOUND` rows to the authors.
- **Reading lists and theses:** find retracted sources before you cite them.

### FAQ

**Does NOT_FOUND prove a reference is fake?** No. It means none of the four indexes holds a matching paper. Books, theses, reports, conference talks and very recent papers are often missing. Check each one before accusing anyone.

**Why FOUND but with YEAR_DIFFERS?** The paper exists with the same title and authors, but the cited year is not within a year of the indexed one. Preprint and journal years often differ; large gaps deserve a look.

**Does it read PDFs?** No. Paste the reference list as text.

### Honest limits

- Matching is heuristic: it looks for the indexed title and first author inside your reference text, so badly formatted references can be marked `MISMATCHED`.
- Retraction data is only as current as Crossref, OpenAlex and PubMed.
- Only the reference is checked, not whether the paper supports the claim it is cited for.

### Price

Pay per event: **$0.002 per reference checked** (`reference-checked`, charged when a source gave a definite answer) and **$0.01 per bibliography report** (`paper-report`). A 60-reference paper costs about **$0.13**.

# Actor input Schema

## `references` (type: `string`):

Paste your reference list: plain text (one reference per line or numbered), BibTeX entries, or a list of DOIs. Do not mix BibTeX and plain text in one box. Leave everything empty to run a built-in demo.

## `dois` (type: `array`):

Optional list of DOIs or doi.org links, one per line. Each is checked for existence and retraction status.

## `contactEmail` (type: `string`):

Optional. Crossref and OpenAlex give faster, more reliable service to callers that identify themselves. Leave empty to stay anonymous.

## `checkRetractions` (type: `boolean`):

Flag retracted papers, expressions of concern and corrections using Crossref (Retraction Watch data), OpenAlex and PubMed publication types.

## `useOpenAlex` (type: `boolean`):

Use OpenAlex as a second index when Crossref has no match. Wider coverage of preprints and books, but slightly slower.

## `usePubMed` (type: `boolean`):

Use PubMed as a last resort for biomedical references that Crossref and OpenAlex cannot match.

## `onlyProblems` (type: `boolean`):

Output only references that are not found, mismatched, retracted, flagged with a concern, or carry a year difference. Clean references are then not charged.

## `maxReferences` (type: `integer`):

Upper bound on how many references are checked in one run (hard limit 1000).

## `concurrency` (type: `integer`):

How many references are looked up in parallel. Keep it low to stay inside the public APIs' rate limits.

## Actor input object example

```json
{
  "references": "",
  "dois": [],
  "contactEmail": "",
  "checkRetractions": true,
  "useOpenAlex": true,
  "usePubMed": true,
  "onlyProblems": false,
  "maxReferences": 200,
  "concurrency": 3
}
```

# Actor output Schema

## `results` (type: `string`):

No description

## `downloadCsv` (type: `string`):

No description

## `summary` (type: `string`):

No description

## `count` (type: `string`):

No description

# API

You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup.

## JavaScript example

```javascript
import { ApifyClient } from 'apify-client';

// Initialize the ApifyClient with your Apify API token
// Replace the '<YOUR_API_TOKEN>' with your token
const client = new ApifyClient({
    token: '<YOUR_API_TOKEN>',
});

// Prepare Actor input
const input = {
    "references": "",
    "dois": [],
    "contactEmail": "",
    "checkRetractions": true,
    "useOpenAlex": true,
    "usePubMed": true,
    "onlyProblems": false,
    "maxReferences": 200,
    "concurrency": 3
};

// Run the Actor and wait for it to finish
const run = await client.actor("that_red_bird/fake-reference-checker").call(input);

// Fetch and print Actor results from the run's dataset (if any)
console.log('Results from dataset');
console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`);
const { items } = await client.dataset(run.defaultDatasetId).listItems();
items.forEach((item) => {
    console.dir(item);
});

// 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs

```

## Python example

```python
from apify_client import ApifyClient

# Initialize the ApifyClient with your Apify API token
# Replace '<YOUR_API_TOKEN>' with your token.
client = ApifyClient("<YOUR_API_TOKEN>")

# Prepare the Actor input
run_input = {
    "references": "",
    "dois": [],
    "contactEmail": "",
    "checkRetractions": True,
    "useOpenAlex": True,
    "usePubMed": True,
    "onlyProblems": False,
    "maxReferences": 200,
    "concurrency": 3,
}

# Run the Actor and wait for it to finish
run = client.actor("that_red_bird/fake-reference-checker").call(run_input=run_input)

# Fetch and print Actor results from the run's dataset (if there are any)
print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}")
for item in client.dataset(run.default_dataset_id).iterate_items():
    print(item)

# 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start

```

## CLI example

```bash
echo '{
  "references": "",
  "dois": [],
  "contactEmail": "",
  "checkRetractions": true,
  "useOpenAlex": true,
  "usePubMed": true,
  "onlyProblems": false,
  "maxReferences": 200,
  "concurrency": 3
}' |
apify call that_red_bird/fake-reference-checker --silent --output-dataset

```

## MCP server setup

```json
{
    "mcpServers": {
        "apify": {
            "type": "http",
            "url": "https://mcp.apify.com/?tools=fetch-actor-details,that_red_bird/fake-reference-checker"
        }
    }
}
```

The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer <APIFY_API_TOKEN>` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations).

## OpenAPI specification

Download the OpenAPI definition: https://api.apify.com/v2/actors/yNcQelCYd6oGxOBWF/builds/Dob4YoBfs40fS55d6/openapi.json