Entropy Box / Notes

Technical notes

Things we measured, things we got wrong, and rules that turned out to generalise. Written to be useful independently of our system — if a note is only interesting to someone who already cares about Entropy Box, it does not belong here.

01  Embedding similarity cannot decide duplication
We used 1,155 adjudicated near-duplicate pairs from our own registry as labelled data. Once the gate has flagged a pair, precision is 0.069 and AUC is 0.629 — and raising the threshold does not repair it. Why a conservative adjudication stage is a requirement rather than a safety margin, and why we could only run the experiment because the decisions had been logged with their reasons.
negative result · entity resolution · dataset published
02  Merge on reasoning, split on artifacts
Two multi-agent architectures failed before the third worked. The rule that explains both failures: merge two stages into one context when what passes between them is reasoning; split them when what passes between them is an artifact. Multi-agent decomposition is usually justified by capability specialisation — we think the binding constraint is what survives the handoff.
multi-agent design · context engineering · two failures

Part of Entropy Box, a knowledge compiler for embodied AI. Posts are CC BY 4.0 — republish them, with attribution and a link back. Corrections: wangyuqi11and11@163.com.