[Save this search]

Status
All
   Fixed (7187)
  Closed (5171)
✓Open (2647)
   Won't Fix (545)
   Duplicate (297)
   Invalid (217)
   Not A Problem (195)
Issue type
✓All
  Issue (2280)
  PR (367)
Author relation
✓All
  None (2064)
  Member (663)
  Contributor (505)
  New contributor (67)
Created
✓All
  Past day (2)
  Past 2 days (5)
  Past 3 days (6)
  Past week (25)
  Past month (54)
  Past 3 months (135)
  Past 6 months (198)
  Past year (300)
Updated
✓All
  Past day (14)
  Past 2 days (15)
  Past 3 days (17)
  Past week (47)
  Past month (107)
  Past 3 months (199)
  Past 6 months (248)
  Past year (550)
Updated ago
✓All
  > 1 day ago (2633)
  > 2 days ago (2632)
  > 3 days ago (2630)
  > 1 week ago (2600)
  > 1 month ago (2540)
  > 3 months ago (2448)
  > 1 year ago (2097)
Comment count
✓All
  0 (585)
  1 (376)
  2 - 5 (822)
  6 - 10 (448)
  10 - 20 (316)
  > 20 (174)
Reaction count
✓All
  0 (2360)
  1 (174)
  2 - 5 (91)
  6 - 10 (16)
  10 - 20 (5)
  > 20 (1)
Review Requested
✓All
  jpountz (13)
  mikemccand (12)
  rmuir (6)
  iverase (4)
  benwtrent (4)
  dweiss (3)
  msokolov (3)

See all 26...
Mentioned
✓All
  jpountz (99)
  mikemccand (84)
  rmuir (63)
  benwtrent (53)
  uschindler (42)
  msokolov (40)
  romseygeek (22)

See all 211...
Reviewed
✓All
  mikemccand (26)
  jpountz (23)
  rmuir (15)
  dweiss (14)
  benwtrent (14)
  uschindler (13)
  msokolov (12)

See all 67...
Commented
✓All
  asfimport (1368)
  github-actions[bot] (276)
  jpountz (137)
  mikemccand (127)
  rmuir (97)
  msokolov (87)
  benwtrent (82)

See all 311...
User
✓All
  asfimport (1810)
  github-actions[bot] (334)
  mikemccand (218)
  jpountz (207)
  rmuir (141)
  benwtrent (121)
  msokolov (109)

See all 470...
Last comment user
✓All
  asfimport (1319)
  github-actions[bot] (268)
  mikemccand (35)
  jpountz (31)
  rmuir (23)
  msokolov (18)
  benwtrent (17)

See all 167...
Draft
✓All
  No (305)
  Yes (62)
Component
✓All
  core (643)
  analysis (155)
  highlighter (46)
  facet (42)
  spatial (41)
  queryparser (29)
  sandbox (23)

See all 23...
Type
✓All
  enhancement (1203)
  bug (726)
  task (200)
  test (78)
  documentation (23)
Labels
✓All
  Stale (264)
  legacy-jira-fix-versio... (214)
  legacy-jira-fix-versio... (169)
  affects-version:4.0-ALPHA (81)
  tool:build (54)
  vector-based-search (48)
  affects-version:6.0 (37)

See all 157...
Commits?
✓All
  No (2647)
Reporter
✓All
  rmuir (270)
  mikemccand (154)
  jpountz (124)
  dsmiley (70)
  uschindler (53)
  iverase (44)
  dweiss (42)

See all 762...
Assignee
✓All
  Unassigned (2406)
  mikemccand (35)
  uschindler (32)
  romseygeek (27)
  dsmiley (22)
  rmuir (15)
  jpountz (11)

See all 44...
  
Filters: Status (Open),  Issue type,  Author relation,  Created,  Updated,  Updated ago,  Comment count,  Reaction count,  Review Requested,  Mentioned,  Reviewed,  Commented,  User,  Last comment user,  Draft,  Component,  Type,  Labels,  Commits?,  Reporter,  Assignee

#16030 PR: Implement data-blind scalar quantization
12.7 minutes ago  6 comments  0 votes  0 watches  github-actions[bot], kaivalnp, mccullocht, mikemccand, romseygeek
Add an option to the quantization format to enable or disable centering (enabled by default). When centering is disabled we also stop writing the float vectors which can lead to significant ... Special handling is included during merges -- we check that all of the input is in the same ...
    github-actions[bot] 4.1 days ago:  This PR has not had activity in the past 2 weeks, labeling it as stale. If the PR is waiting for review, notify the dev@lucene.apache.org list. Thank you for your contribution!
    mccullocht 18.5 days ago:  I'll try to revive this over the next week or two. > This should work well when incoming vectors are isotropic (all dimensions behave the same i.e ...

#16029: Data-blind scalar quantization
25.4 minutes ago  8 comments  0 votes  0 watches  Pranshu-S, mccullocht, mikemccand, romseygeek, shubhamvishu, xande
Description Add options for data-blind scalar quantization. Data blind quantization will allow users to drop the raw float vectors which will be a significant ... There are 3 parts to this change: - [ ] Allow users to disable centering and drop the raw float ...
    mccullocht 12.5 days ago:  We had a similar discussion about rotation and centering on a PR for TurboQuant. Quantizers often perform better with a Gaussian component distribution -- RaBitQ and TurboQuant ... If you don't _need_ to be data blind, centering is still tremendously helpful.
    Pranshu-S 12.7 days ago:  Right — the RaBitQ-inspired prototype I was referring to does use a centroid/residual ... My mention of changing the estimator wasn't intended as an extension of the data-blind ... I had gone down a separate path while trying to understand the remaining low-bit scoring loss after ...

#16506 PR: Add a scalar-quantized version of the de-duplicating vector format
1.2 hours ago  50 comments  0 votes  0 watches  github-actions[bot], kaivalnp, msokolov
Description Add a scalar-quantized version of the de-duplicating vector format DedupHnswScalarQua ... De-duplication works like the raw format: vectors of the same dimension and encoding form a group, ... Files: - .vdd: Raw vectors and per-field data (same as the flat de-duplicating format)
    kaivalnp 1.2 hours ago:  FYI there was a bug in luceneutil that caused the baseline exact KNN results to be incorrect for ... See https://github.com/mikemccand/luceneutil/pull/614 for the bug and fix. index-time-filter runs ... I re-ran a full benchmark for both main and this PR on 100K Cohere v3 vectors, 1024 dimensions, 10K ...
    kaivalnp 4.6 days ago:  Thanks for the review @msokolov! I'll merge this in a couple of days unless there are further comments.

#16581 PR: Allow updating of single valued SortedNumericDocValues
7.4 hours ago  2 comments  0 votes  0 watches  AdityaTeltia, HoustonPutman, github-actions[bot]
Description As mentioned in 16580, there are good reasons to want to use SortedNumericDocValues ... This PR adds support for updating docValues when SortedNumericDocValues fields are actually ... There's very little meat here, most of it is just allowing the NumericDocValues update logic to be ...
    AdityaTeltia 1.9 days ago:  This rejection happens at write time, but writeFieldUpdates only prunes pendingDVUpdates after ... The catch block rethrows and the pruning loop below it never runs, so the offending DocValuesFieldUp ... Every subsequent commit, NRT reopen, merge, and RAM-triggered flush re-enters this loop and throws ...

#16578 PR: Add Query#isKnownToMatchAllDocs() to replace == checks
8.8 hours ago  2 comments  0 votes  0 watches  github-actions[bot], michaeljmarshall, msokolov
Description I put this together as a draft to see what it'd look like to replace the == ... It seems valuable for AbstractKnnVectorQuery in the case where a MatchAllDocsQuery is wrapped in a ... However, what is not clear is if the other == MatchAllDocsQuery.class checks can correctly use ...
    michaeljmarshall 8.8 hours ago:  > We expect Query classes that match all docs to rewrite to MatchAllDocsQuery.INSTANCE to indicate ... For the following code, we get new BoostQuery(new ConstantScoreQuery(MatchAllDocsQuery.INSTANCE), 0 ... Should I open a PR to add the following: `java case FILTER: if ( ...
    msokolov 1.7 days ago:  This seems like adding an extra abstraction, but why do we need it? We expect Query classes that match all docs to rewrite to MatchAllDocsQuery.INSTANCE to indicate ... If they comply with this expectation, there's no need for a new method that indicates the same thing

#16540 PR: Store doc values disi data into a separate file
9.5 hours ago  2 comments  0 votes  0 watches  costin, github-actions[bot], sgup432
Description This moves out the doc values disi data (used to store sparse data) into a separate ... Closes - https://github.com/apache/lucene/issues/16298 <!-- If this is your first contribution to Lucene, please make sure you have reviewed the ...
    sgup432 9.5 hours ago:  > wanted to check if this was a conscious tradeoff and if there's any data assessing the impact. Yeah it was a conscious choice as I took the idea mentioned here(https://github.com/apache/lucene/ ... In a nutshell, the idea was to make the layout format more flexible and the reader can use the ...
    costin 4.6 days ago:  +1 on moving IndexedISI into its own file. A few things stood out to me: The writeDISI method buffers the entire IndexedDISI blob via BBDO ... For typical fields this is fine however for large segments, the bitmap could end up to a few MB; ...

#16585 PR: docs: Add SECURITY.md for Apache Lucene
9.8 hours ago  22 comments  0 votes  0 watches  dweiss, javanna, msokolov, rmuir, sgup432, uschindler
As the number of reported security issues also went up - and many of them were rejected, I used ... I am not sure if we should also add this to our web page, I'd like to start with the Git repo. The text was mocked by Claude Fable 5 (Claude Max provided for ASF committers) and I rewrote + ...
    uschindler 9.8 hours ago:  > "the application embedding it" ~~ "the application in which it is embedded" is how I was finally ... It is obvious if you read the previous sentence.
    sgup432 9.9 hours ago:  Looks better now IMO

[15.8 msec search, 16.8 msec total]