Verified65% confidenceFactExact time
The core mechanism of BM25 is based on three factors: term frequency (TF), inverse document frequency (IDF), and document length normalization, with parameter k1 typically set between 1.2 and 2.0 and b typically set at 0.75.
3
Sources
65%
Confidence
Long-term
Relevance
9/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe BM25 score consists of three parts: IDF, term-frequency saturation (k1), and length normalization (b), with the IDF formula being log((N - n + 0.5) / (n + 0.5) + 1).72% similarUnverifiedUsing a 4B model to rerank the candidate shortlist filtered by BM25 outperformed BM25 alone (+0.047, confidence interval [0.020, 0.073]), even though the 4B reranker itself did not pass qualification.68% similarUnverified生产中典型做法是先用向量检索和BM25快速召回50-100个候选结果，再用reranker精排取top-k送入LLM67% similarUnverifiedVector retrieval returns cosine similarity or distance scores, while BM25 returns term-frequency-based scores; the two have different units, so raw scores cannot be directly added or compared.67% similarUnverifiedUsing the Qwen3-27B model on the wikitext-2 dataset with fp32 reference logits as a baseline for KL divergence measurement, the median KLD dropped from 0.0023 to 0.000001, an approximately 2300-fold improvement in precision.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/848729API
curl https://kongchang.com/api/v1/knowledge/claims/848729MCP
get_claim(id=848729)