The tokenizer read every long word backwards (V-664)

encodeWord backtracks the Viterbi path from the end of the word and
prepends each piece, which puts them back in reading order. A second
reverse after that loop undid it. So "query: вода" tokenized to
[0 12 1294 41 12489 2] where the reference tokenizer gives
[0 41 1294 12 12489 2], and every multi-piece Russian word reached the
model with its pieces in the wrong order.

Measured on the recall fixture, same 27 cases either way:

  recall@1  70.4% -> 77.8%
  recall@3  85.2% -> 96.3%
  answered after gate  63.0% -> 66.7%
  false recall  0/5 -> 1/5

The classifier barely moves, 76.0% to 75.0% on the routing fixture,
because seeds and queries were mangled the same way and cosine survived
it. Recall is where it cost, because a stored passage and a live query
are different lengths and break differently.

The embedder id now names a tokenizer revision. Stored vectors were
written under rev 1 and no longer sit in the same space as a query
embedded now, and the model file's name never moved, so nothing would
have triggered ReembedAll.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013ptwopxyo3Z2kwFckHkLvN
This commit is contained in:
2026-08-08 22:23:56 +04:00
parent 50c6637c1b
commit feabf9f350
2 changed files with 30 additions and 10 deletions
+12 -2
View File
@@ -1,10 +1,13 @@
package router
import "testing"
import (
"strings"
"testing"
)
func TestEmbedderIDFromModelPath(t *testing.T) {
got := modelIDFromPath("/opt/maven/models/embedder/multilingual-e5-small.onnx")
if got != "multilingual-e5-small@384" {
if got != "multilingual-e5-small@384/tok2" {
t.Fatalf("modelIDFromPath = %q", got)
}
// A different model file must produce a different id, even at 384 dim.
@@ -12,6 +15,13 @@ func TestEmbedderIDFromModelPath(t *testing.T) {
if old == got {
t.Fatal("two different models share one id")
}
// The tokenizer is half of what makes a vector, and it changes under a
// model file whose name never moves (V-664). An id that ignored it would
// leave stored passages in one space and every new query in another, with
// nothing to trigger the re-embed.
if !strings.Contains(got, "/tok") {
t.Fatalf("id %q does not name the tokenizer revision", got)
}
}
func TestEmbedderIDIncludesDim(t *testing.T) {