| # Evaluation operations |
|
|
| Use the commit-pinned participant data and the same code revision as participants. |
| Validate all submissions before semantic scoring. Current scoring works without |
| a reference file; optional public evaluation assets record reference availability |
| and the matching task hash, not 50 new answers. |
|
|
| ```bash |
| tib download --organizer --output evaluation |
| tib judge --data data/participant --submissions runs/agent/submissions \ |
| --base-url "$JUDGE_BASE_URL" --model "$JUDGE_MODEL" \ |
| --audit-documents 160 --output runs/agent/reviews |
| tib evaluate --data data/participant --submissions runs/agent/submissions \ |
| --reviews runs/agent/reviews --output runs/agent/report.json |
| ``` |
|
|
| Set JUDGE_API_KEY locally. Nonempty tasks incur claim-blind annotation requests |
| (at most 12 sampled documents per batch, up to four concurrent requests), followed |
| by one quality request. A malformed blind response permits one repair request. |
| The default 160-document audit needs at least 14 blind requests before quality |
| grading; long texts can require more. Abstentions make no API calls. |
| No reference-matching requests occur for current |
| tasks. Configure model, timeout, max input characters and output tokens; an |
| oversized packet fails explicitly without silent truncation. Use a sufficiently |
| large model context or adjust the documented audit budget (60–2000 documents). |
| Many quotations may require raising a very small audit budget. |
|
|
| Persist judge_config.json for the audit seed and configuration, and each task's |
| .blind.json for the bound, reusable pre-grading annotations. Completed valid |
| reviews are reused. Changed tasks, corpora, submissions or judge configuration |
| require fresh output directories. Do not tune an agent on judge feedback and |
| present that score as a frozen run. Publish all coverage and uncertainty metrics, |
| not only a favorable subset. |
| |
| The structural verifier is exhaustive; semantic review is sampled and fallible. |
| This release does not require independent verification or add an unseen holdout. |
| For enforceable comparisons, run participants in an isolated environment without |
| evaluation access; the process adapter alone cannot enforce this. |
| |
