CORD-v2 test receipt · read by Qwen3.5-4B (QLoRA) with best-of-4 and the accounting checker
Receipt photo

How the checker chose
The model wrote one normal reading and four sampled ones. The checker keeps the reading that passes the most accounting rules; the model's confidence only breaks ties.
- Greedy50% rulespicked
- Sample 150% rules
- Sample 250% rules
- Sample 350% rules
- Sample 450% rules
Extracted fields
| Item | Qty | Unit price | Amount | Remove |
|---|---|---|---|---|
Accounting checks
goes to human review
- Items add up to the subtotalexpected 95,000 · found 57,000
- Cash − change = totalexpected 57,000 · found 57,000
The rules recheck as you edit, using a TypeScript port of the Python checker that agrees with it on all 277 rule results of the test set. Amounts within 1% are treated as equal, to allow for rounding.