claude
2ea39a3d41
Search Kiwix for the topic, not the whole sentence (V-668)
...
Kiwix ranks by keyword overlap, which the package doc has said since it was
written: "why is the sky blue" finds a TV episode. queryKiwix sent the whole
Russian sentence, because the verbatim path added by V-508 skips the rewriter
that would have reduced it.
Measured against the Russian ZIM on 2026-08-09, over eight questions. Four
reach the right article where they did not: TCP was "Перехват TCP-соединения"
and is TCP, фотосинтез was "C4-фотосинтез" and is Фотосинтез, Линус Торвальдс
was "Tux", and "кто написал Войну и мир" was "Радуйся, мир (Доктор Кто)".
Two were already right and stay right. Two are still wrong and were wrong
before. Nothing regressed.
kiwix.Topic drops the narrative request, the interrogative and a verb behind
one, and keeps everything else. A word it cannot classify is more likely the
topic than noise. TitlePath tries the exact article first, since a ZIM is
addressable by title and a wrong title is a 404.
The gate this task set out to build does not exist. Query-to-passage cosine
scored 0.79-0.91 on answerable questions and 0.75-0.84 on unanswerable ones,
and the sets overlap. The wrong TCP article scored 0.8653, above five of six
unanswerable rows. e5 measures topic, not whether the passage answers.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com >
Claude-Session: https://claude.ai/code/session_013ptwopxyo3Z2kwFckHkLvN
2026-08-09 10:42:45 +04:00
..
2026-07-03 00:42:35 +02:00
2026-08-06 02:58:37 +04:00
2026-08-01 21:43:28 +04:00
2026-08-04 01:26:52 +04:00
2026-08-04 17:56:39 +04:00
2026-08-06 20:55:53 +04:00
2026-08-05 02:32:03 +04:00
2026-08-04 17:56:39 +04:00
2026-08-05 18:43:13 +04:00
2026-08-01 14:16:56 +04:00
2026-08-01 14:16:56 +04:00
2026-08-06 01:39:38 +04:00
2026-08-01 22:32:37 +04:00
2026-08-03 22:57:11 +04:00
2026-08-02 02:24:36 +04:00
2026-08-09 10:42:45 +04:00
2026-08-06 04:11:47 +04:00
2026-08-05 19:53:47 +04:00
2026-08-05 19:54:21 +04:00
2026-08-05 13:32:24 +04:00
2026-08-05 02:27:22 +04:00
2026-08-05 12:07:07 +04:00
2026-08-04 03:44:48 +04:00
2026-08-04 19:10:00 +04:00
2026-08-06 21:54:13 +02:00
2026-08-06 21:54:13 +02:00
2026-08-05 21:40:15 +04:00
2026-08-02 02:28:53 +04:00
2026-08-04 00:33:30 +04:00
2026-08-08 22:24:47 +04:00
2026-08-08 22:24:47 +04:00
2026-08-08 21:38:32 +04:00
2026-08-04 04:04:59 +04:00
2026-08-06 01:23:37 +04:00
2026-08-01 05:49:27 +04:00
2026-08-06 01:05:50 +04:00
2026-08-06 05:23:31 +04:00
2026-08-05 13:32:24 +04:00
2026-08-01 22:21:28 +04:00
2026-08-04 16:47:23 +04:00
2026-08-06 03:12:59 +04:00
2026-08-06 05:12:20 +04:00
2026-08-08 22:24:47 +04:00
2026-08-06 20:12:19 +04:00
2026-08-08 22:24:47 +04:00
2026-08-05 13:32:24 +04:00
2026-07-31 23:09:22 +04:00
2026-08-06 11:47:27 +04:00
2026-08-01 16:29:56 +04:00
2026-08-04 16:00:17 +04:00
2026-08-06 02:48:00 +04:00
2026-08-06 05:23:31 +04:00
2026-08-04 18:01:28 +04:00
2026-08-04 16:00:17 +04:00
2026-08-06 02:48:00 +04:00
2026-08-04 18:12:25 +04:00
2026-08-04 18:12:25 +04:00
2026-08-01 14:15:33 +04:00
2026-07-30 23:49:22 +04:00
2026-08-08 22:24:47 +04:00
2026-07-20 12:00:37 +04:00
2026-08-07 00:32:54 +04:00
2026-08-08 22:24:47 +04:00
2026-08-03 13:40:33 +04:00
2026-08-05 13:11:04 +04:00
2026-08-04 20:46:53 +04:00
2026-08-01 05:08:08 +04:00
2026-08-05 15:10:35 +04:00
2026-08-05 15:10:35 +04:00
2026-08-05 20:54:11 +04:00
2026-08-05 20:54:11 +04:00
2026-08-01 14:10:36 +04:00
2026-08-01 14:10:36 +04:00
2026-08-01 14:05:13 +04:00
2026-08-01 14:05:13 +04:00
2026-08-05 15:52:54 +04:00
2026-08-01 19:46:39 +04:00
2026-08-05 15:52:54 +04:00
2026-07-31 02:15:30 +04:00
2026-08-01 14:05:07 +04:00
2026-08-01 14:20:04 +04:00
2026-08-06 21:54:13 +02:00
2026-08-01 14:11:57 +04:00
2026-08-01 14:11:57 +04:00
2026-08-02 02:16:45 +04:00
2026-08-03 10:12:21 +02:00
2026-08-01 14:36:48 +04:00
2026-08-04 03:17:29 +04:00
2026-08-04 16:41:11 +04:00
2026-08-04 15:37:54 +04:00
2026-08-08 22:24:47 +04:00
2026-08-06 01:39:38 +04:00
2026-08-05 18:16:54 +04:00
2026-08-05 18:16:54 +04:00
2026-08-02 09:58:07 +04:00
2026-08-01 01:30:06 +04:00
2026-08-04 04:35:42 +04:00
2026-08-04 04:35:42 +04:00
2026-08-05 22:36:10 +04:00
2026-08-05 22:36:10 +04:00
2026-08-06 02:49:09 +04:00
2026-08-05 13:11:04 +04:00
2026-08-05 01:31:23 +04:00
2026-08-06 00:56:29 +04:00
2026-08-05 15:28:26 +04:00
2026-08-05 15:28:26 +04:00
2026-08-09 01:56:46 +04:00
2026-08-01 21:35:58 +04:00
2026-08-04 18:45:52 +04:00
2026-08-08 22:24:47 +04:00
2026-07-31 13:30:38 +04:00
2026-08-04 20:46:53 +04:00
2026-08-05 01:48:18 +04:00
2026-08-04 03:53:20 +04:00
2026-08-04 21:18:52 +04:00
2026-08-06 02:58:49 +04:00
2026-08-06 20:12:19 +04:00
2026-08-06 20:12:19 +04:00
2026-08-06 21:11:42 +02:00
2026-08-06 21:11:42 +02:00
2026-08-06 19:45:37 +04:00
2026-08-06 20:12:19 +04:00
2026-08-06 04:43:38 +04:00
2026-08-06 04:43:38 +04:00
2026-08-02 02:24:36 +04:00
2026-08-05 01:10:17 +04:00
2026-08-05 01:10:17 +04:00
2026-08-05 22:59:56 +04:00
2026-08-05 23:05:09 +04:00
2026-08-08 22:24:47 +04:00
2026-08-02 09:35:47 +04:00
2026-08-01 14:06:38 +04:00
2026-08-04 15:54:13 +04:00
2026-08-01 21:40:06 +04:00
2026-08-04 18:45:52 +04:00
2026-08-01 14:12:43 +04:00
2026-08-01 14:12:43 +04:00
2026-07-31 23:27:08 +04:00
2026-08-09 00:42:15 +04:00
2026-07-31 14:14:18 +04:00
2026-08-06 20:53:48 +04:00
2026-08-06 19:45:37 +04:00
2026-08-04 20:46:53 +04:00
2026-08-04 05:33:00 +04:00
2026-08-04 05:33:00 +04:00
2026-08-05 02:27:22 +04:00
2026-08-06 02:08:05 +04:00
2026-08-05 22:59:56 +04:00
2026-08-05 22:59:56 +04:00
2026-08-08 21:43:15 +04:00
2026-08-08 21:43:15 +04:00
2026-08-06 02:20:03 +04:00
2026-08-01 14:21:46 +04:00
2026-08-01 14:21:46 +04:00
2026-08-06 21:11:42 +02:00
2026-08-08 23:21:51 +02:00
2026-08-06 03:33:01 +04:00
2026-08-06 03:33:01 +04:00
2026-08-05 20:51:46 +04:00
2026-08-04 01:19:41 +04:00
2026-08-04 01:19:41 +04:00