Add rerank module: bi-encoder retrieval plus an ONNX Runtime cross-encoder reranker behind a Spring AI DocumentPostProcessor; hit@k and MRR by pool size and question kind, latency by candidates, batch size, threads and int8, and five reproduced traps. Local MiniLM models, no database or API key.

Co-Authored-By: Claude Sonnet 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_01Cfo2K4u4rF18QsPXPFnytf
2026-10-24 18:00:05 +00:00
parent 2ccf1d8d19
commit a429aa40f6
34 changed files with 1617 additions and 0 deletions
@@ -0,0 +1,6 @@
+# A (query, passage) pair as the cross-encoder receives it
+
+model inputs: [input_ids, attention_mask, token_type_ids]
+tokens        : [[CLS], what, is, e, -, 412, ##6, ?, [SEP], error, e, -, 412, ##6, -, ref, ##und, window, closed, ., ref, ##unds, can, be, issued, for, 120, days, ., [SEP]]
+token_type_ids: [0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
+attention_mask: [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]