Struct SparseQueryConfig 

pub struct SparseQueryConfig {
    pub tokenizer: Option<String>,
    pub weighting: QueryWeighting,
    pub heap_factor: f32,
    pub weight_threshold: f32,
    pub max_query_dims: Option<usize>,
    pub pruning: Option<f32>,
    pub min_query_dims: usize,
    pub lsp_gamma: Option<usize>,
}

Query-time configuration for sparse vectors

Quality-sensitive query optimization knobs. Weight filtering, dimension caps, fractional pruning, finite LSP gamma, and heap factors below 1.0 can all change the candidate set. They are disabled by default and should be tuned against representative Recall@K or relevance judgments.

Fields

tokenizer: Option<String>

HuggingFace tokenizer path/name for query-time tokenization Example: “Alibaba-NLP/gte-Qwen2-1.5B-instruct”

weighting: QueryWeighting

Weighting strategy for tokenized query terms

heap_factor: f32

Heap factor for approximate search (SEISMIC-style optimization) A block is skipped if its max possible score < heap_factor * threshold

• 1.0 = exact search (default)
• values below 1.0 = increasingly aggressive block pruning

weight_threshold: f32

Minimum weight for query dimensions (query-time pruning) Dimensions with abs(weight) below this threshold are dropped before search. Useful for filtering low-IDF tokens that add latency without improving relevance.

• 0.0 = no filtering (default)
• positive values drop dimensions and require quality validation

max_query_dims: Option<usize>

Maximum number of query dimensions to process (query pruning) Processes only the top-k dimensions by weight

• None = process all dimensions (default, exact)
• Some(k) = process only the top-k dimensions by absolute weight

pruning: Option<f32>

Fraction of query dimensions to keep (0.0-1.0), same semantics as indexing-time pruning: sort by abs(weight) descending and keep the top fraction. BMP uses this subset for candidate generation and the bounded full query for final scoring; MaxScore uses the subset for both. None or 1.0 = no pruning.

min_query_dims: usize

Minimum number of query dimensions before pruning and weight_threshold filtering are applied. Protects short queries from losing most signal.

Default: 4. Set to 0 to always apply pruning/filtering.

lsp_gamma: Option<usize>

LSP/0 top-superblock guarantee γ. None selects the paper-derived schedule from retrieval depth; Some(0) requests exhaustive traversal.

Trait Implementations

impl Clone for SparseQueryConfig

fn clone(&self) -> SparseQueryConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SparseQueryConfig

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

Formats the value using the given formatter.

impl Default for SparseQueryConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for SparseQueryConfig

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for SparseQueryConfig

fn eq(&self, other: &SparseQueryConfig) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl Serialize for SparseQueryConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for SparseQueryConfig