Skip to content

Dev #14

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
zTgx merged 2 commits into from
Apr 5, 2026
Merged

Dev #14

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
23a4802
feat: add BM25 scoring functionality with configurable parameters
zTgx Apr 5, 2026
52cfbf1
chore(release): bump version from 0.1.15 to 0.1.16
zTgx Apr 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
[package]
name = "vectorless"
version = "0.1.15"
version = "0.1.16"
edition = "2024"
authors = ["zTgx <beautifularea@gmail.com>"]
description = "Hierarchical, reasoning-native document intelligence engine"
Expand Down Expand Up @@ -80,6 +80,9 @@ roxmltree = "0.20"
# Random number generation (for sampling)
rand = "0.8"

# BM25 scoring
bm25 = { version = "2.3.2", features = ["parallelism"] }

[dev-dependencies]
tempfile = "3.10"
tokio-test = "0.4"
Expand Down
165 changes: 10 additions & 155 deletions src/retrieval/content/scorer.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@
use std::collections::HashMap;

use crate::document::NodeId;
use crate::retrieval::search::{extract_keywords, Bm25Params, STOPWORDS};
use crate::util::estimate_tokens;

use super::config::ScoringStrategyConfig;
Expand Down Expand Up @@ -130,8 +131,7 @@ pub struct RelevanceScorer {
/// Scoring strategy to use.
strategy: ScoringStrategyConfig,
/// BM25 parameters.
k1: f32,
b: f32,
params: Bm25Params,
}

impl RelevanceScorer {
Expand All @@ -142,8 +142,7 @@ impl RelevanceScorer {
Self {
query_keywords,
strategy,
k1: 1.2,
b: 0.75,
params: Bm25Params::default(),
}
}

Expand All @@ -153,8 +152,7 @@ impl RelevanceScorer {
Self {
query_keywords: keywords,
strategy,
k1: 1.2,
b: 0.75,
params: Bm25Params::default(),
}
}

Expand Down Expand Up @@ -240,13 +238,15 @@ impl RelevanceScorer {
continue;
}

// IDF calculation
// IDF calculation using BM25L variant
let df = ctx.doc_freq.get(&term_lower).copied().unwrap_or(1) as f32;
let idf = ((ctx.doc_count as f32 - df + 0.5) / (df + 0.5) + 1.0).ln();

// BM25 formula
let numerator = tf * (self.k1 + 1.0);
let denominator = tf + self.k1 * (1.0 - self.b + self.b * doc_len / ctx.avg_doc_len);
let k1 = self.params.k1;
let b = self.params.b;
let numerator = tf * (k1 + 1.0);
let denominator = tf + k1 * (1.0 - b + b * doc_len / ctx.avg_doc_len);

score += idf * numerator / denominator;
}
Expand All @@ -263,159 +263,14 @@ impl RelevanceScorer {
}
}

/// Extract keywords from a query string.
fn extract_keywords(query: &str) -> Vec<String> {
// Common English stop words
const STOPWORDS: &[&str] = &[
"a",
"an",
"the",
"is",
"are",
"was",
"were",
"be",
"been",
"being",
"have",
"has",
"had",
"do",
"does",
"did",
"will",
"would",
"could",
"should",
"may",
"might",
"must",
"shall",
"can",
"need",
"dare",
"ought",
"used",
"to",
"of",
"in",
"for",
"on",
"with",
"at",
"by",
"from",
"as",
"into",
"through",
"during",
"before",
"after",
"above",
"below",
"between",
"under",
"again",
"further",
"then",
"once",
"here",
"there",
"when",
"where",
"why",
"how",
"all",
"each",
"few",
"more",
"most",
"other",
"some",
"such",
"no",
"nor",
"not",
"only",
"own",
"same",
"so",
"than",
"too",
"very",
"just",
"and",
"but",
"if",
"or",
"because",
"until",
"while",
"about",
"what",
"which",
"who",
"whom",
"this",
"that",
"these",
"those",
"i",
"me",
"my",
"myself",
"we",
"our",
"ours",
"ourselves",
"you",
"your",
"yours",
"yourself",
"yourselves",
"he",
"him",
"his",
"himself",
"she",
"her",
"hers",
"herself",
"it",
"its",
"itself",
"they",
"them",
"their",
"theirs",
"themselves",
];

query
.to_lowercase()
.split(|c: char| !c.is_alphanumeric())
.filter(|s| {
let s = *s;
!s.is_empty() && s.len() > 1 && !STOPWORDS.contains(&s)
})
.map(String::from)
.collect()
}

/// Compute information density of content.
fn compute_density(content: &str) -> f32 {
let words: Vec<&str> = content.split_whitespace().collect();
if words.is_empty() {
return 0.0;
}

// Stopword ratio (lower is better)
const STOPWORDS: &[&str] = &[
"a", "an", "the", "is", "are", "was", "were", "be", "been", "being", "have", "has", "had",
"do", "does", "did", "will", "would", "could", "should", "may", "might", "must", "shall",
"can", "to", "of", "in", "for", "on", "with", "at", "by", "from", "and", "but", "or", "as",
];

// Use shared STOPWORDS from bm25 module
let stopword_count = words
.iter()
.filter(|w| STOPWORDS.contains(&w.to_lowercase().as_str()))
Expand Down
Loading