3 posts tagged with "enterprise-ai"

Milvus Hybrid Retrieval: Weighted Fusion Scores vs. Similarity Thresholds

March 8, 2026 · 3 min read

Debugging hybrid retrieval scoring in a RAG knowledge-base project — the full troubleshooting trail below.

TL;DR

Milvus hybrid retrieval with weighted fusion scores as 0.7 * dense_score + 0.3 * sparse_score tops out around 0.7 in theory — and lower in practice. Filtering with min_similarity=0.7 removes nearly everything. Fix: drop the threshold to 0.3, or adapt it to the fusion strategy dynamically.

RAG Query Sources Missing the similarity Field

March 8, 2026 · 3 min read

CCLEE

Debugging the query-result response format in a RAG knowledge-base project — the full troubleshooting trail below.

TL;DR

The RAG /query endpoint returned sources containing metadata only, with no per-source similarity score. Fix: merge metadatas and distances while assembling the response, computing similarity = 1 - distance for each source.

RAG Knowledge Base and AI Floating Widget Solution

February 11, 2026 · 2 min read

TL;DR​

TL;DR​

TL;DR

TL;DR