Crate frankensearch_rerank 

FlashRank cross-encoder reranking for frankensearch.

Provides the FlashRankReranker implementation of the Reranker trait, using ONNX Runtime for cross-encoder scoring with sigmoid activation on raw logits. Gracefully falls back to original scores if the model is unavailable.

Model Layout

Required files in the model directory:

• onnx/model.onnx (preferred) OR model.onnx (legacy)
• tokenizer.json

Architecture

Cross-encoders differ from bi-encoders: instead of comparing pre-computed embeddings, they process the query and document together through a transformer, producing direct token-level attention between them. This is dramatically more accurate but cannot pre-compute anything.

(query, document) → tokenize → ONNX → logit → sigmoid → score ∈ [0, 1]

Re-exports

pub use pipeline::DEFAULT_MIN_CANDIDATES;

pub use pipeline::DEFAULT_TOP_K_RERANK;

pub use pipeline::rerank_step;

Modules

pipeline

Rerank pipeline step: integrates cross-encoder reranking into the search pipeline.

Structs

FlashRankReranker

FlashRank cross-encoder reranker backed by ONNX Runtime.

Constants

DEFAULT_MAX_LENGTH

Default maximum input token length for cross-encoder pairs.

DEFAULT_MODEL_NAME

Default model directory name for the FlashRank nano cross-encoder.

Functions

find_model_dir

Search for a FlashRank model directory in standard locations.