Woody a7a22f1494 fix(relevance): tolerate LLM score count mismatches via padding instead of discarding ... The per-sub-question filter was all-or-nothing: if the LLM returned 9 scores for 10 chunks (common with qwen3.5-35b), every chunk was discarded and the user got 'no relevant information found'. Now: fewer scores → pad with 0.0; more scores → truncate. Changed from error→warning since this is recoverable. Also improve LTT page UI: sources collapsed by default in per-sub-q sections, and the 'Your question' text now shows the full question instead of being truncated. Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent) Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>		2026-04-27 14:31:18 +08:00
..
__init__.py	feat: Phase 1.1 project setup with config, database, and models	2026-04-22 16:13:52 +08:00
embedding_client.py	feat(backend): add embedding client and update LLM client	2026-04-23 13:26:43 +08:00
history_service.py	feat(db): update history schema and generate prompt template for Package 4	2026-04-26 23:28:28 +08:00
llm_client.py	debug(backend): add LLM request/response logging for OpenRouter debugging	2026-04-23 16:28:43 +08:00
prompt_service.py	feat(prompts): integrate filter_per_subq with PromptService, fix seed bugs, restructure UI	2026-04-27 11:14:27 +08:00
query_decomposer.py	feat(history): Phase 3.5 — Query History backend (service, API, timing, XML capture)	2026-04-25 22:59:53 +08:00
rag.py	feat(services): add per-sub-question retrieval, filtering, and response generation	2026-04-26 23:27:50 +08:00
relevance_filter.py	fix(relevance): tolerate LLM score count mismatches via padding instead of discarding	2026-04-27 14:31:18 +08:00