glam/backend
kempersc 99dc608826 Refactor RAG to template-based SPARQL generation
Major architectural changes based on Formica et al. (2023) research:
- Add TemplateClassifier for deterministic SPARQL template matching
- Add SlotExtractor with synonym resolution for slot values
- Add TemplateInstantiator using Jinja2 for query rendering
- Refactor dspy_heritage_rag.py to use template system
- Update main.py with streamlined pipeline
- Fix semantic_router.py ordering issues
- Add comprehensive metrics tracking

Template-based approach achieves 65% precision vs 10% LLM-only
per Formica et al. research on SPARQL generation.
2026-01-07 22:04:43 +01:00
..
ducklake feat: add web archives integration and improve enrichment scripts 2025-12-07 17:49:07 +01:00
postgres remove a,bihguous web-claims 2025-12-21 00:01:54 +01:00
rag Refactor RAG to template-based SPARQL generation 2026-01-07 22:04:43 +01:00
sync enrich all custodian timespan 2025-12-15 22:31:41 +01:00
typedb enrich all custodian timespan 2025-12-15 22:31:41 +01:00
valkey update postgis data 2025-12-10 23:51:51 +01:00