Skip to content

Current evaluation result

Immutable evidence. This page is the result page generated inside the checksum-verified v0.3.4 publication bundle, at source revision e9ffdf26a2a216c4c1fcc0dc810b0b498e726abe. It is reproduced here without retyping a score; the evidence map records its manifest and report checksums.

Partition: evaluation only. These tables do not include or pool development cases.

  • Freeze ID: real-project-v2
  • Bounded claim: This independent slice supports only descriptive per-profile comparisons for the source-only sampled Java, Rust, and C++ repositories and the References and Definition operations recorded here. Report v2 separately. A combined v1/v2 view may preserve and aggregate only documented language/profile strata and their denominators; it must not pool the slices into language-wide, ecosystem-wide, cross-language, latency, memory, or cold-start claims.
  • Audited corpus: 12 repositories and 36 cases
LanguageReference profileRegistered adapterRepositoriesCasesPopulation exclusionsSource-review replacements
javaeclipse-jdtlsadapters/lsp/eclipse-jdtls.json41227 (repository has fewer than 20 eligible source files × 3; repository lacks a GitHub-reported SPDX license × 10; repository lacks root build marker for java × 10; repository reported source size exceeds 157286400 bytes × 17; repository source tree is truncated by GitHub × 2)none
rustrust-analyzeradapters/lsp/rust-analyzer.json41240 (repository has fewer than 20 eligible source files × 3; repository lacks a GitHub-reported SPDX license × 11; repository lacks root build marker for rust × 9; repository reported source size exceeds 157286400 bytes × 29; repository source tree is truncated by GitHub × 3)none
cppapple-clangd-21adapters/lsp/apple-clangd-21.json41236 (repository has fewer than 20 eligible source files × 1; repository lacks a GitHub-reported SPDX license × 13; repository lacks root build marker for cpp × 11; repository reported source size exceeds 157286400 bytes × 22; repository source tree is truncated by GitHub × 1)none
ArtifactFileSHA-256
Protocolbenchmarks/evaluation/real-project-v2/protocol.json1c0ff8a5c0fc4f4a66b564dcdb9dd384e1b9f5ee28134a1dd34770b85a4ffbf1
Selectionbenchmarks/evaluation/real-project-v2/selection.json88a853512223c5c4859df2c9fe068e83364a3c3bb84defdd2d9f2c502aa58b2c
Independent reviewbenchmarks/evaluation/real-project-v2/review.jsona22a32fa5945fc3d716b866f3fb9b4b055fce9984534dd8e2417fae569b86e1d
Source lockbenchmarks/evaluation/real-project-v2/sources.jsoncd0684b7210d63ae21b7b953699e53e1eaafd750ec8918bb4ae783dc9a377857
Reviewer openai-gpt-5.6-solbenchmarks/evaluation/real-project-v2/reviews/openai-gpt-5.6-sol.json956a043ea74e0e894d156c45f842e16b6d7cb3e9cdb0274ef57406c45e179afe
Reviewer anthropic-claude-fable-5benchmarks/evaluation/real-project-v2/reviews/anthropic-claude-fable-5.json47d53e24dccf31a7b8c06f746f0d338bca1bcb38d2f9ae3810ce6b56effccd99
Adjudication real-project-v2-human-adjudicationbenchmarks/evaluation/real-project-v2/reviews/adjudication.json726076c0759eb88a836185f9430dc1ff4f6fb7fbe67d9d7fdabe686423bc59db
CandidateRunnerRequested versionProfileEnvironmentReport SHA-256
apple-clangd-21clangd21.0.0adapters/lsp/apple-clangd-21.jsonmacos/“aarch64”/Native/HostSpecific2e8167d21745617e6d507875bb190206d30954d5ee5f2ad873db4825ff514c3f
bifrostbifrost8ddf13625653caf927137a0de1966401055debad—macos/“aarch64”/Native/HostSpecific0341a8f8aee26d5f8ccd88b48f1cd665efc2a49ed914418346ccd7227c62a03e
eclipse-jdtlsJDT Language Server (Standard)1.61.0-202607142124adapters/lsp/eclipse-jdtls.jsonmacos/“aarch64”/Native/HostSpecificb0b41090ff1d6f1f4393761148c181800926bb09eb213ca91e2d3bc17a45b2ca
rust-analyzerrust-analyzer2026-07-13adapters/lsp/rust-analyzer.jsonmacos/“aarch64”/Native/HostSpecificb039a8be300568355668808de8a8161ae1e2735a770395aeb47d8e55753be2a4
Reference profileSharedBifrost foundReference found
Apple clangd 21.0.0128/12 (66.7%)1/12 (8.3%)
Eclipse JDT LS 1.61.0-2026071421241212/12 (100.0%)6/12 (50.0%)
rust-analyzer 2026-07-131211/12 (91.7%)12/12 (100.0%)

TP, FP, and FN are reported without true negatives. Strict precision counts every extra result; policy-adjusted precision excludes authored and policy-allowed extras.

Reference profileAnalyzerCasesTPFPFNDestination recallExact-token recallStrict precisionPolicy-adjusted precisionExact-set case rateExtras/success
Apple clangd 21.0.0Bifrost121430994.1%94.1%100.0%100.0%66.7%0.00
Apple clangd 21.0.0Apple clangd126049239.5%39.5%93.8%93.8%8.3%0.00
Eclipse JDT LS 1.61.0-202607142124Bifrost127500100.0%100.0%100.0%100.0%100.0%0.00
Eclipse JDT LS 1.61.0-202607142124Eclipse JDT LS121805724.0%24.0%85.7%100.0%33.3%0.25
rust-analyzer 2026-07-13Bifrost12730198.6%98.6%100.0%100.0%91.7%0.00
rust-analyzer 2026-07-13rust-analyzer1274240100.0%100.0%71.2%75.5%58.3%1.25
Reference profileSharedBoth exactBifrost onlyReference onlyNeither
Apple clangd 21.0.0121704
Eclipse JDT LS 1.61.0-202607142124126600
rust-analyzer 2026-07-13127401