cybermaggedon 01cc8dbc64 feat: replace LLM edge scoring with cross-encoder reranker in GraphRAG (#1005) ... Replace the three-prompt LLM scoring pipeline (kg-edge-scoring, kg-edge-reasoning, kg-edge-selection) with a cross-encoder reranker service backed by FlashRank. The new hop_and_filter() method performs iterative graph traversal with semantic scoring at each hop, replacing the previous follow_edges/get_subgraph approach. - Add reranker service (trustgraph-base client/service, FlashRank processor) - Add gateway dispatch for reranker via API and WebSocket - Rewrite GraphRAG pipeline: hop_and_filter() with per-hop cross-encoder scoring - Remove kg_prompt() and edge_score_limit from prompt client - Update provenance: add tg:EdgeSelection type, tg:concept, tg:score predicates - Update CLIs (tg-invoke-graph-rag, tg-show-explain-trace) for new metadata - Add tg-invoke-reranker CLI tool - Add tech spec and UX developer guidance - Update all unit and integration tests		2026-06-30 14:36:37 +01:00
..
__init__.py	Updated test suite for explainability & provenance (#696)	2026-03-13 14:27:42 +00:00
test_agent_provenance.py	Additional agent DAG tests (#750)	2026-04-01 13:59:58 +01:00
test_dag_structure.py	feat: replace LLM edge scoring with cross-encoder reranker in GraphRAG (#1005)	2026-06-30 14:36:37 +01:00
test_explainability.py	Additional agent DAG tests (#750)	2026-04-01 13:59:58 +01:00
test_graph_rag_chain.py	Forward missing explain_triples through RAG clients and agent tool callback (#768)	2026-04-08 11:41:17 +01:00
test_triples.py	Add agent explainability instrumentation and unify envelope field naming (#795)	2026-04-13 16:16:42 +01:00
test_uris.py	Use UUID-based URNs for page and chunk IDs (#703)	2026-03-21 21:17:03 +00:00
test_vocabulary.py	Updated test suite for explainability & provenance (#696)	2026-03-13 14:27:42 +00:00