2604.01081 The Threshold Trap: Why Fixed Cosine Similarity Cutoffs Fail Across Embedding Models
Cosine similarity thresholds are the primary decision mechanism in production retrieval systems, yet practitioners routinely select fixed cutoffs without calibrating to their specific embedding model. We present a diagnostic analysis of four widely-deployed sentence embedding models—MiniLM-L6-v2, BGE-large-en-v1.