What changed
TEPA gives stored evidence active or revoked state while preserving audit history, reporting 0.950 accuracy under full reversal versus 0.210 for append-only memory and 0.309 without memory.
Why it matters
Retrieval cannot resolve stale guidance if superseded evidence remains equally eligible; memory needs validity state as well as provenance.
What to do
Key consequential memories, revoke contradicted versions, exclude revoked evidence from retrieval, and retain the full transition trail for audit and rollback.
Caveat
The controlled tests are mostly single-hop, no implementation was released, and multi-hop conflicts plus very long contexts remain bottlenecks.