k-NN over the corpus by embedding
emem_find_similarRead-onlyIdempotent
Find similar places by cell embedding or inline vector. Returns k nearest neighbors using cosine, Hamming, or combined similarity.
Instructions
k-NN over the corpus by cell embedding or inline vector.
When to use: Call when the user asks 'find places like X', 'where else looks like this', or hands an embedding to find neighbours. key is either a cell64 or inline:[x,y,...]. Default band is geotessera (128-D Tessera foundation embedding); pass band: "geotessera.multi_year" for the 1152-D 9-vintage (2017–2025) fusion.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| k | No | ||
| key | Yes | cell64 (look up that cell's vector) or 'inline:[x,y,...]' literal vector | |
| band | No | vector band to scan (default: 128-D Tessera foundation embedding). For mode=hamming/hamming_then_rerank you can pass either the cosine band (e.g. 'geotessera') or its binary sibling ('geotessera.bin128') — the responder picks the right one. | geotessera |
| mode | No | Scoring mode. cosine = fp32 over full vector (precise, ~256 B/cell scan). hamming = sign-bit popcount over the binary sibling band (~16 B/cell, ~1000× faster, ~65% recall@10). hamming_then_rerank = triage with Hamming on 4·k candidates then re-rank by cosine — matches cosine precision at ~16× less work. | cosine |
| as_of_tslot | No | Bi-temporal valid-time bound. Applied to candidate cells BEFORE cosine scoring — a cell with no fact whose tslot ≤ as_of_tslot under the scoring band is dropped from the candidate pool (undecidable→drop). When set, the Lance ANN fast-path is bypassed (the index has no signed_at column); brute-force k-NN runs instead so as_of is honoured truthfully. | |
| as_of_signed_at | No | Bi-temporal transaction-time bound (RFC 3339). Also applied to candidates BEFORE cosine. Same Lance-bypass note as as_of_tslot. |