Vector · semantic similarity
Does this assertion mean something close to the query? Compatible query embeddings search the LanceDB assertion projection.
MESA V4 · RETRIEVAL
Each lane proposes eligible candidates for a different reason. Fusion combines their ranks; provenance keeps the reasons inspectable.
Does this assertion mean something close to the query? Compatible query embeddings search the LanceDB assertion projection.
Does the evidence text contain useful terms or a scoped citation? FTS5 ranks passages; exact citation lookup contributes within the lane.
Do subject, predicate, entity/literal object and evidence match? SQLite-owned relations provide a separate signal.
Is there a bounded relationship path supported by eligible canonical assertions? Paths retain supporting IDs and direction.
RECIPROCAL RANK FUSION
RRF(candidate) = Σ weight_lane / (k + rank_lane)
k = 60 · vector = 8.0 · BM25/assertion/graph = 1.0 · rank starts at 1
A duplicate candidate casts one vote per lane. Ties resolve by candidate identity. A lane may return no eligible hit. This example uses one first-ranked candidate; it does not simulate the complete retrieval engine.
Citation boosts apply inside lanes: BM25 exact +200, same statute +40; assertion exact +150, same statute +40. They affect lane order, not RRF increments. When MESA_LEGAL_RULES_ENABLED=true, a query-dependent legal factor after fusion is 1.5 for exact statute/article, 1.25 for the same statute, 0.25 for a competing statute, and 1.0 otherwise.
Responses expose rrf_score, legal_factor, raw_score and final_score separately. Higher fused relevance is better; none of these fields is a probability of truth.
With legal rules enabled, an exact law/article or a decision’s own docket/decision reference reserves at most three leading positions in document order after fusion and optional reranking. Decision identity must start within the first 300 characters of chunk zero; later quotations do not count. Other hits retain their order. Scoped titles and first-chunk law headers teach law names, including K.; AYM alone does not select the Constitution.
Versioned legal_items retain original chunk offsets separately from SPO facts and do not add fusion score. MESA_LEGAL_RULES_ENABLED=false bypasses legal processing and hides stored legal overlays in returned copies while preserving generic retrieval and durable snapshots.
Legal settings and limitsAt most 20 seeds, 3 hops and 500 requested graph hits; frontier 64, per-node fan-out 32, default request timeout 15 seconds. PageRank is telemetry only. These limits do not guarantee latency or completeness.
The default 2048-token budget counts formatted model-visible evidence with cl100k_base. If BPE cannot load, a conservative UTF-8 byte upper bound may fit less. Complete graph proof fits atomically.
Single query is the default. Adaptive mode triggers on empty Q0 or no two-lane support in its top three; at most one planner and two expansions keep the same scope. Outer fusion weights are Q0=1.0 and Q1/Q2=0.5. Failure preserves Q0.
Capability flags describe operational support. An expected failure of a configured vector/graph backend returns explicit 503. Projection-only graph adapters do not advertise neighbor traversal.
Vector weight defaults to 8.0; use MESA_RRF_VECTOR_WEIGHT=1 for the previous setting. Zero removes a lane contribution. Optional BGE reranking is off by default, needs [ml] and adds rerank_score after fusion while preserving rrf_score and final_score.
Read the DEV matrix, CPU/GPU latency and limits →FOR DEVELOPERS · CONTRACT TEST
Core has a deterministic contract fixture: one “Alice knows Aurora” assertion ranks first in all four lanes, with RRF = 4/61 under explicit unit weights (the runtime vector default is 8.0). It uses real SQLite and Kùzu with mock vectors; it is not a quality benchmark.
“Alice”AliceknowsAurora
One assertion. Its source still attached.
source-chunk-1doc-1 / chunk-1test_real_four_lanes_fuse_once_and_repeat_deterministicallyCore test in the v0.9.3 release.
Source baseline: Core v0.9.3.