Skip to content
Clairos Learn

Retrieval and embeddings · Lesson 9 of 16 · 9 min

Spot a missing citation

A citation must connect a claim to source evidence that actually supports it. Merely including a document identifier somewhere in an answer is not adequate grounding.

Mastery 0/3Run itExplain itApply it
learner@clairosSimulated AI Ops lab
learner@clairos:~ $
↑ history · Tab complete · Ctrl+C cancel · Esc, Tab to leaveExit 0

Understand the pattern

A citation must connect a claim to source evidence that actually supports it. Merely including a document identifier somewhere in an answer is not adequate grounding.

Interpret the evidence

Correct the claim to match the source or withhold it; the citation does not support 60 days.

A tempting mistake

A cited document makes any nearby numeric claim valid.

All names, measurements, prices, versions, and thresholds in this exercise are fictional training data. The evidence supports this bounded scenario, not a claim about a live provider.

Worked example and expected result
$ cat /opt/ai-lab/cases/aiops-retrieval-embeddings-009.json
{
  "fixture": "Fictional training data; not a live measurement.",
  "claim": "refunds allowed for 60 days",
  "source_text": "refunds allowed for 30 days",
  "claim_supported": false
}

Exit status: 0. The command completed successfully in the training fixture.

$ jq .claim_supported /opt/ai-lab/cases/aiops-retrieval-embeddings-009.json
false

Exit status: 0. The command completed successfully in the training fixture.

$ printf '%s\n' 'correct-unsupported-claim' > /home/learner/lab/aiops-retrieval-embeddings-009-decision.txt
(no output; exit 0)

Exit status: 0. The command completed successfully in the training fixture.

Try a variation

Name the assumption that would have to change for “correct-unsupported-claim” to stop being the justified action. Do not silently replace missing evidence with a guess.

Explain it

What does the evidence in “Spot a missing citation” justify?

Apply it elsewhere

A colleague concludes: “A cited document makes any nearby numeric claim valid.” What should you do?

Keep the momentum.

Complete the lab, then explain and apply the pattern.

Next: Diagnose empty retrievalTry an incident drill

Anonymous lesson and mastery progress lasts for this session. Sign in to keep synced lesson XP between sessions; mastery stays device-local. No speed bonus, lost lives, or hint penalties.

Keep learning

  1. 01Compare similarity scoresSimilarity scores are ranking signals under a specific embedding and metric setup. They are not calibrated probabilities of truth or correctness.Practice
  2. 02Diagnose empty retrievalEmpty retrieval can come from filters, missing documents or an unavailable index. Inspect those boundaries before loosening access restrictions or inventing an answer.Practice
  3. 03Inspect duplicate chunksDuplicate chunks can crowd a result list and reduce source diversity. Review stable document/chunk identifiers before deciding whether two similar passages are redundant.Practice