부모 노트: Personal LLM Memory Lab. 이전 평가: Architecture Evaluation — 2026-07-15
Decision
The next Personal LLM Memory Goal is stage-gated:
- Phase A — Evidence-Preserving Typed Memory Transition Compiler v1
- Phase B — Fingerprint-Keyed Persistent Sparse/Graph Cache, activated only after Phase A passes its promotion gates.
D1 retrieval remains fixed during Phase A so compiler improvements can be attributed to the write path. Markdown remains the only durable knowledge store.
Why this Goal
The completed leading pair failed 49 of 80 held-out cases: 47 at compilation and two at query. K3 passed 63.75% of compiler hard gates, reached 42.5% exact plans, and showed weak equivalent-plan and minimal-delta consistency. Production still lacks a typed semantic plan/apply core, while D1 rebuilds disposable retrieval state per request.
Candidate comparison
- Typed evidence-preserving K3 core — 91: Phase A
- Persistent sparse/graph index — 73: Phase B after compiler promotion
- Answer-level evaluation — 49: Subsequent Goal
- Longitudinal operations — 48: Subsequent Goal
- Registered dense Q3 — 38: Optional later experiment
The fixed weights cover observed bottleneck impact, prerequisite fit, independent measurability, privacy and reversibility, maintainability, scale, token/context leverage, and operational simplicity. The totals are planning support, not benchmark performance.
Phase A promotion boundary
- Production-owned typed source, claim, temporal, provenance, plan, action, precondition, query-effect, and review-item contracts.
- Explicit
ADD,CONFIRM,CONTEST,SUPERSEDE,RETRACT,FORGET, andNOOPtransitions. - Canonical fingerprints, dry-run default, guarded atomic apply, replay safety, rollback evidence, and unchanged
raw/. - A new independently frozen 100-sequence suite: 20 development and 80 one-shot held-out sequences. The consumed suite is never rerun.
- At least 90% exact transition/end-state accuracy, at least 95% action precision and recall, and perfect hard gates for idempotency, provenance, authorized forgetting closure, raw immutability, privacy, unrelated-state preservation, and zero-write no-op.
- No new forbidden D1 retrieval and no greater than two-percentage-point registered compatibility regression.
Phase B promotion boundary
- Cache identity includes corpus fingerprint, policy scope, schema/profile version, and tokenizer/ranker version.
- Stale, missing, corrupt, or interrupted state triggers canonical fallback or atomic rebuild.
- Exact D1 parity for paths, ordering, answerability, claim states, and bounded context.
- Crash recovery, concurrent-reader safety, invalidation, atomic replacement, and secure forgetting tests.
- Measured full/incremental build, consistency delay, p50/p95 latency, context and byte/token proxies, memory, storage, and operator steps at small, 5,000-unit, and 50,000-unit scales.
Explicit deferrals
Dense Q3, answer-model judging, months-long effectiveness, MCP-first implementation, autonomous graph rewriting, reinforcement learning, and a vendor database are not success substitutes for this Goal.
Internal source gap
Workspace research found only the existing Personal LLM Memory Lab and Architecture Evaluation — 2026-07-15. They provide prior-result context but no independent new evidence for candidate selection.
Primary and official research
- TrustMem
- HaluMem
- MemoryAgentBench
- Memora
- Agent Memory systems characterization
- EvoMemBench
- AtomMem
- Useful Memories Become Faulty When Continuously Updated by LLMs
- Zep/Graphiti and official deletion semantics
- W3C PROV-O
- RFC 6902 JSON Patch
- RFC 8785 JCS
- JSON Schema Draft 2020-12
- SQLite FTS5
- FaithJudge
Local durable record
The complete provenance, scoring rationale, counterevidence, and Goal contract are stored in the local Wiki as wiki/syntheses/personal-memory-next-goal-decision.md, backed by immutable source note raw/research/2026-07-15-personal-memory-next-goal-selection.md.
These local paths were not present in the llm-wiki tree at migration time. This Notion report is kept as the remaining copy.
Status: decision recorded; Goal not yet claimed complete.