Domain-specific fine-tuning with a dual MNRL plus cosine-similarity loss improves syllabus-based question answering and narrows the gap to proprietary embeddings.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
An Open-Source Dual-Loss Embedding Model for Semantic Retrieval in Higher Education
Domain-specific fine-tuning with a dual MNRL plus cosine-similarity loss improves syllabus-based question answering and narrows the gap to proprietary embeddings.