IndriDirichletSimilarity(float mu) uses LMSimilarity's default (cf+1)/(|C|+1) collection model instead of IndriCollectionModel (cf/|C|), so new IndriDirichletSimilarity(2000f) scores differently from new IndriDirichletSimilarity() although both print the same name

Finding LFV-744Area: search / similarities (IndriDirichletSimilarity constructors and the LMSimilarity collection model)Priority: lowImpact: silent wrong resultsNovelty: new

Finding page for LFV-744

AI-assisted finding, confirmed by automated checks, not yet triaged by a human Lucene committer. This site files nothing upstream.

What it shows

You install new IndriDirichletSimilarity(2000f) and expect the same similarity as new IndriDirichletSimilarity(), which is documented as using 'the default mu value of 2000', and Indri's Dirichlet smoothing log((tf + mu*cf/|C|)/(|D| + mu)) for any mu. Instead the mu-only constructor passes no collection model, so its implicit super() installs LMSimilarity's DefaultCollectionModel, (cf+1)/(|C|+1), while only the no-argument constructor installs the class's own IndriCollectionModel, cf/|C|. On a 25-token index the term 'president' (cf 3) scores -1.871553 instead of -2.119101 in the same document, although both instances print 'LM IndriDirichlet(2000.000000)'.

The test shows that new IndriDirichletSimilarity(2000f) scores differently from new IndriDirichletSimilarity() (documented as mu = 2000), because the mu-only constructor falls back to LMSimilarity's (cf+1)/(|C|+1) collection model instead of IndriCollectionModel's cf/|C|.

Expected vs actual

Expected equal explain() scores for term a in the one-document index a b; actual expected:<-0.6931472> but was:<-0.4057149> (-ea and -da).

Files

• src/org/apache/lucene/search/similarities/TestIndriDirichletMuConstructor.java (JUnit test)

Run it

• Tested on Lucene main e1faeba08bde (2026-11-02): still reproduces.
• Tested on Lucene main 645d4fe14a20 (2026-10-28): reproduced.
• Recorded outcome of run 1 on Lucene main 645d4fe14a20: 1 test(s) run, 1 failed.
• How to read the outcome: a failing test is the finding reproducing (the failure message states expected against actual).
• Seeds: a rerun with the same seed replays the same random choices. Assertions off (-da -dsa -Dtests.asserts=false) is the production-like setting.

# Lucene checkout at the tested commit, built with: ./gradlew --no-daemon assemble   (JDK 25)
LUCENE=/path/to/lucene; V=12.0.0-SNAPSHOT
CP=$LUCENE/lucene/core/build/libs/lucene-core-$V.jar
CP=$CP:$LUCENE/lucene/codecs/build/libs/lucene-codecs-$V.jar
CP=$CP:$LUCENE/lucene/test-framework/build/libs/lucene-test-framework-$V.jar
CP=$CP:randomizedtesting-runner-2.9.2.jar:junit-4.13.2.jar:hamcrest-3.0.jar   # test-framework dependencies (Maven Central)
CP=$CP:junit-platform-console-standalone-6.1.3.jar   # JUnit Platform console launcher (Maven Central)

javac -d classes -cp "$CP" $(find src -name '*.java')

# run 1: assertions on, seed DEADBEEF
java -ea -esa --add-modules jdk.incubator.vector --enable-native-access=ALL-UNNAMED -Dtests.seed=DEADBEEF \
     -cp "classes:$CP" org.junit.platform.console.ConsoleLauncher \
     execute --select-class org.apache.lucene.search.similarities.TestIndriDirichletMuConstructor

Gradle alternative: copy the test into the matching module's src/test/ tree of the checkout and run ./gradlew -p lucene/<module> test --tests <class> -Ptests.seed=<seed> (add -Ptests.asserts=false for a run without assertions).

Feedback

Open an issue in lucene-ai-bug-hunt. Please do not open issues or pull requests in apache/lucene about this finding unless a Lucene committer agrees it should be filed.

Snapshot of 2026-11-04. AI-assisted; not yet triaged by a human Lucene committer. Feedback: open an issue in lucene-ai-bug-hunt.