Marius CiepluchandClaude Opus 5.5 da91ef0dd9 exp-003: supplementary all-units replay after the cache fill; exp-004 registered ...

Supplementary analysis, not a new decision: exp-003's verdict stands.
The 142 units excluded for an uncached response were filled with the same
six incumbent programs, parameters and request keys, on exp-003's
937-example memory (the harvest's 196 rows and its 2026-10-03 embeddings
left out; a zero-cost replay first reproduced exp-003 exactly: 211
decided, 1056 replayed, $0.582876). Fill: MeteredClient, ceiling $0.99
under the approved $1.00, 697 client calls, $0.331759, no provider error.

Replay on all 353 units (244 paired, 146 groups): 0 client calls, $0.0,
1768 responses replayed; no arm flags any unit, the registered rule reads
inconclusive. Calibration holds only for untrusted destination (ECE
0.0454) and arm A still finds no BLOCK point there.
benchmarks/experiments/exp-003-precision-bound-blocking/result-complete-cache.json

exp-004 registered in s3://sast-memory (manifest 87fceb65..., pin
2ad9c1d, units pending): benchmarks/experiments/exp-004-pooled-block-gate/
registration.json. docs/decision-engine.md: both, briefly.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo

2026-10-03 09:47:20 +02:00

..

exp-001-known-callers

bench(learn): exp-001 and exp-002 registered in the memory store, units pending (decision-engine task 7.2)

2026-10-02 16:08:36 +02:00

exp-002-retrieval-ensemble

bench(learn): exp-002 result -- retrieval ensemble vs temperature sampling

2026-10-02 19:05:34 +02:00

exp-003-precision-bound-blocking

exp-003: supplementary all-units replay after the cache fill; exp-004 registered

2026-10-03 09:47:20 +02:00

exp-004-pooled-block-gate

exp-003: supplementary all-units replay after the cache fill; exp-004 registered

2026-10-03 09:47:20 +02:00