Glossary
Short definitions for the terms that show up in Jev requests and answers.
Short definitions for the terms that show up in Jev requests and answers.
Abstention is a first-class outcome: low confidence, other, or a Noul band near 0.5 — your code refuses to auto-act.
Action space is the closed set of acts your code may take after Jev answers — including abstain. Not a generated plan.
Calibration means higher confidence should mean higher accuracy. TypeSafe claims RLCD trains for this; you still measure it.
Choice picks one option from a map of up to 255 criteria and returns a distribution plus confidence.
Confidence is a 0–1 summary of how peaked a Choice or Score distribution is. Absent on Noul.
A confidence threshold is your act/review/abstain floor on Choice or Score confidence — illustrations in docs are not defaults.
A decision contract is the closed questions, criteria, and code thresholds you ship — not a TypeSafe legal term.
A decision trace is the logged state hash, questions, answers, model id, and policy branch — not a generated chain of thought.
In jev.pro usage, a guardrail is a typed Jev question plus your threshold — not a certified security product.
Intent is a closed label you ask Jev to assign so code can route to a handler — not a generated plan.
jev-latest is the SDK-default alias for the current official Jev release. It can move.
Noul is Jev's yes/no primitive: a probability in [0, 1] with no separate confidence field.
A ranked route is an ordered list of handlers from Jev probabilities — code still applies floors before it follows rank 1.
RLCD is TypeSafe's name for Reinforcement Learning for Calibrated Decisions.
Score places a state on 2–10 ordered levels and returns a possibly fractional expected position.
State is the text or JSON payload every question in a request evaluates.
System One is TypeSafe's class of models for fast structured decisions. Jev is the first.
Define audit trail in the Jev decision model — Jev audit trail definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define evidence bundle in the Jev decision model — Jev evidence bundle definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define fallback in the Jev decision model — Jev fallback definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define human handoff in the Jev decision model — Jev human handoff definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define policy check in the Jev decision model — Jev policy check definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define retrieval grounding in the Jev decision model — Jev retrieval grounding definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define risk band in the Jev decision model — Jev risk band definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define schema validation in the Jev decision model — Jev schema validation definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define signal in the Jev decision model — Jev signal definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define threshold in the Jev decision model — Jev threshold definition. Independent unofficial notes; confirm against docs.typesafe.ai.
Define verdict in the Jev decision model — Jev verdict definition. Independent unofficial notes; confirm against docs.typesafe.ai.