task
Frozen task
The question, source set, baseline, and scoring rule are fixed before the run is interpreted.
frontiers / frontier
AI-for-science benchmark state
- id
- vfr_efc649fd772a1ff1
- license
- CC-BY-4.0
- findings
- 12
- accepted core
- 12
- contested
- 0
- links
- 0
- sources
- 1
- evidence
- 12
- avg conf
- 0.30
e24/24 · finding.noted · reviewer:will-blair · 2026-06-10 · 6c12→d02f
attempt
Attempt packet
The agent or reviewer, capability, system, input material, declared output material, environment, and failures stay attached to the result.
evaluation
Measured result
The evaluation record pins the target, outcome, score, evidence refs, evaluator, and timestamp.
review
Accepted meaning
Review decides what the result means for the frontier. The score is evidence, not the event.
No benchmark runs
No frozen-task evaluations on record.
ver_* records live under .vela/evaluations/.