Spaces:
Sleeping
Sleeping
Commit History
-
Fix: Pass user query to LLM context in Fast Lane and Deep Dive stages
7f34cf9 -
Upgrade to Hybrid Architecture v3 (Fast Lane + Deep Dive)
e2d9873
-
Remove conversation history from LLM context to reduce tokens
7b3dbf5 -
Add compact prompts and context builder - reduces token usage by ~50%
cb3e125 -
Fix: pass context to _generate_response for persona/zone/trend/weather
a6e4432
-
Full Developer Spec implementation: Priority-based Multi-Stage Reasoning
- models.py: Pydantic schemas for all data structures - prompts.py: Enhanced LLM prompts with JSON output formats - context_aggregator.py: Priority-based satellite data formatting - reasoning_engine.py: 4-stage pipeline with evidence chains - priority_mapper.py: Intent-to-context mapping per spec - intent_classifier.py: Enhanced with phrases and regional support - app.py: Spec-compliant response with suggested_followupsee8289e -
Implement multi-stage reasoning architecture (Claim→Validate→Contradict→Confirm)
e2fa5a7