Struct RetrievalCandidate 

pub struct RetrievalCandidate {
    pub id: Uuid,
    pub vector_score: Option<f64>,
    pub text_score: Option<f64>,
    pub graph_distance: Option<u32>,
    pub rrf_score: Option<f64>,
}

Pre-computed retrieval signals for a single candidate entity.

All fields are Option — a missing signal scores 0.0. The runtime layer is responsible for populating whichever fields are available before handing the slice to an objective.

Fields

id: Uuid

Stable entity UUID.

vector_score: Option<f64>

Cosine similarity to the query vector (0.0–1.0).

text_score: Option<f64>

BM25/FTS relevance score (0.0–1.0 normalised, or raw rank score).

graph_distance: Option<u32>

Hop distance from the nearest anchor node (0 = anchor itself).

rrf_score: Option<f64>

Pre-fused RRF score from FusionStrategy::Rrf.

Trait Implementations

impl Clone for RetrievalCandidate

fn clone(&self) -> RetrievalCandidate

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for RetrievalCandidate

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl HasId for RetrievalCandidate

fn id(&self) -> Uuid

Returns the stable UUID identifier for this candidate.

impl Objective<RetrievalCandidate> for VectorSimilarityObjective

fn score( &self, candidate: &RetrievalCandidate, _context: &ObjectiveContext, ) -> f64

Evaluate a single candidate.

fn name(&self) -> &str

Get the name of this objective.

fn precision(&self, _candidate: &T, _context: &ObjectiveContext) -> f64

Precision (inverse variance) of the score estimate; default 1.0 (fully trusted).

fn passes_score(&self, score: f64, context: &ObjectiveContext) -> bool

Check if a score passes the threshold; non-finite scores never pass.

fn passes(&self, candidate: &T, context: &ObjectiveContext) -> bool

Check if a candidate passes the threshold.

fn batch_score( &self, candidates: &[T], context: &ObjectiveContext, ) -> Vec<(usize, f64)>

Score a batch of candidates and return passing (index, score) pairs.

fn select<'a>( &self, candidates: &'a [T], context: &ObjectiveContext, ) -> Vec<Selection<&'a T>>

Select all passing candidates in score-descending order.

fn select_top<'a>( &self, candidates: &'a [T], n: usize, context: &ObjectiveContext, ) -> Vec<Selection<&'a T>>

Select the top N candidates by precision-weighted score.

impl Objective<RetrievalCandidate> for TextRelevanceObjective

fn score( &self, candidate: &RetrievalCandidate, _context: &ObjectiveContext, ) -> f64

Evaluate a single candidate.

fn name(&self) -> &str

Get the name of this objective.

fn precision(&self, _candidate: &T, _context: &ObjectiveContext) -> f64

Precision (inverse variance) of the score estimate; default 1.0 (fully trusted).

fn passes_score(&self, score: f64, context: &ObjectiveContext) -> bool

Check if a score passes the threshold; non-finite scores never pass.

fn passes(&self, candidate: &T, context: &ObjectiveContext) -> bool

Check if a candidate passes the threshold.

fn batch_score( &self, candidates: &[T], context: &ObjectiveContext, ) -> Vec<(usize, f64)>

Score a batch of candidates and return passing (index, score) pairs.

fn select<'a>( &self, candidates: &'a [T], context: &ObjectiveContext, ) -> Vec<Selection<&'a T>>

Select all passing candidates in score-descending order.

fn select_top<'a>( &self, candidates: &'a [T], n: usize, context: &ObjectiveContext, ) -> Vec<Selection<&'a T>>

Select the top N candidates by precision-weighted score.

impl Objective<RetrievalCandidate> for GraphProximityObjective

fn score( &self, candidate: &RetrievalCandidate, _context: &ObjectiveContext, ) -> f64

Evaluate a single candidate.

fn name(&self) -> &str

Get the name of this objective.

fn precision(&self, _candidate: &T, _context: &ObjectiveContext) -> f64

Precision (inverse variance) of the score estimate; default 1.0 (fully trusted).

fn passes_score(&self, score: f64, context: &ObjectiveContext) -> bool

Check if a score passes the threshold; non-finite scores never pass.

fn passes(&self, candidate: &T, context: &ObjectiveContext) -> bool

Check if a candidate passes the threshold.

fn batch_score( &self, candidates: &[T], context: &ObjectiveContext, ) -> Vec<(usize, f64)>

Score a batch of candidates and return passing (index, score) pairs.

fn select<'a>( &self, candidates: &'a [T], context: &ObjectiveContext, ) -> Vec<Selection<&'a T>>

Select all passing candidates in score-descending order.

fn select_top<'a>( &self, candidates: &'a [T], n: usize, context: &ObjectiveContext, ) -> Vec<Selection<&'a T>>

Select the top N candidates by precision-weighted score.

impl Objective<RetrievalCandidate> for RrfFusionObjective

fn score( &self, candidate: &RetrievalCandidate, _context: &ObjectiveContext, ) -> f64

Evaluate a single candidate.

fn name(&self) -> &str

Get the name of this objective.

fn precision(&self, _candidate: &T, _context: &ObjectiveContext) -> f64

Precision (inverse variance) of the score estimate; default 1.0 (fully trusted).

fn passes_score(&self, score: f64, context: &ObjectiveContext) -> bool

Check if a score passes the threshold; non-finite scores never pass.

fn passes(&self, candidate: &T, context: &ObjectiveContext) -> bool

Check if a candidate passes the threshold.

fn batch_score( &self, candidates: &[T], context: &ObjectiveContext, ) -> Vec<(usize, f64)>

Score a batch of candidates and return passing (index, score) pairs.

fn select<'a>( &self, candidates: &'a [T], context: &ObjectiveContext, ) -> Vec<Selection<&'a T>>

Select all passing candidates in score-descending order.

fn select_top<'a>( &self, candidates: &'a [T], n: usize, context: &ObjectiveContext, ) -> Vec<Selection<&'a T>>

Select the top N candidates by precision-weighted score.