TokenPad
Evaluation

Citation Extractor and Validator

Find uncited claims, invalid indices, and sources you paid for and never used.

Your input

171 characters1 lines0 tokensor drop a file

Citation ExtractorExact
0Token change
Token change0no change
Input tokens0what you pasted
Output tokens0what you would send
Token cost of this result
Output tokens0
As input$0.00
× 100K requests$0.00

Everything on this page runs in your browser. Nothing you paste is transmitted, because there is no server here to transmit it to.

Result
 

Citations to sources that do not exist

Models cite [7] when you provided three sources. It is a specific and common failure, it is trivially detectable, and it usually reaches production because nobody checks.

An out-of-range citation should fail closed in your application: show the claim without a link rather than a broken one, and log it as a quality signal.

The unused sources tell you about retrieval

If you retrieve eight chunks and the answer cites two, six chunks were paid for and discarded. That is a retrieval problem showing up as a cost problem.

Consistently low usage means k is too high or your ranking is weak. Reducing k or adding a reranker usually improves both the bill and the answer.

Frequently asked questions

Should every sentence carry a citation?
Every factual claim should. Transitions and framing sentences need not, so use the uncited list as a review queue rather than a hard requirement.
What causes invalid citations?
Usually a prompt that showed a different number of sources than it provided, or an answer written partly from parametric knowledge. Numbering the chunks explicitly in the prompt reduces it substantially.

More evaluation tools