slowbench
RetrievalInferenceEvalsAgentsFindingsSeriesBenchmarksArchive

One failure, read three times

Routing a ticket to a repository

Twenty-eight repositories and a ticket that belongs to one of them. Fusing two rankers lost, then lost again, then won — with nothing about the fusion changed. Read in order, these are the same event from three distances: the mechanism, the verdict I drew from it, and what the verdict was actually a specification for.

1 posts

Where this stands

Held to be true
Picking the right repository is a retrieval problem with almost no published benchmarks, and the obvious signals do not work on it. Absolute similarity scores failed three separate times as a basis for any threshold. Size bias is real and survives the usual normalisations.
What would break it
A similarity threshold that separates good routing results from bad ones on a held-out set.
Still open
Whether concentration replaces similarity as the usable signal. One repository showed clean separation at p 0.016, the cut point was chosen after seeing it, and no second repository has both hits and misses to test on.
  1. 1
    Retrieval11 min

    Fusing two rankers made routing worse, and the reason is structural

    One signal covered every repository, the other covered 17 of 28. Reciprocal rank fusion punished exactly the repositories the second signal could not see.