When Evidence Changes: Versioned Support and Bounded Belief Repair

At fixed total work and source access, does an append-only record that distinguishes valid time from recorded time and tracks claim dependencies improve current, historical, as-known, and downstream-consequence answers over strong overwrite, chronological-history, and newest-recorded baselines, without increasing false invalidation, unhelpful dispute states, or cost?

Proposes a Versioned Support and Dependency Ledger (VSDL) that separates valid time from recorded time, permits an explicit disputed state, and repairs only the affected dependency cone. E18-A is an executed deterministic constructed diagnostic: 36 passing checks and twelve developer fixtures. E18-B and H18-1/H18-2/H18-3 are proposed and untested. No language-model benchmark, model call, human annotation, production change, or deployment.

• Constructed example: a timestamp-only newest-recorded rule returns a late January correction (60) for a March current query whose correct fixture answer is 75. A counterexample over hand-specified records, not model performance.
• Exact check: 24 temporal-query checks hold across all six arrival orders of the three-record example, plus 12 checks of baseline failure, conflict handling, dependency repair, alternative support, and source integrity.

• E18-A Exact deterministic constructed bitemporal and repair diagnostic Executed (diagnostic)
• E18-B Five-arm versioned evidence study (O0/A0/T0/B0/B1) Proposed

• H18-1 Bitemporal selection improves historical-now and as-known accuracy over deterministic newest-recorded selection at the frozen budget.
• H18-2 Dependency-directed repair improves downstream-consequence correctness beyond bitemporal selection alone.
• H18-3 An explicit disputed state reduces unsupported commitment without collapsing supported coverage.

Proposed protocol only. No Noetis behavior, inference, routing, billing, risk value, or collection-gate change.

• Do bitemporal fields beat deterministic newest-recorded assembly on historical and as-known questions at matched cost?
• Does dependency-directed repair improve downstream consequences without excessive false invalidation?
• Does an explicit disputed state reduce unsupported commitment without collapsing useful coverage?

Builds on Round 17, Round 16, Round 15, Round 14, Round 13, Round 10, Round 9. Continued in Round 20, Round 19.