Rethinking Hybrid Retrieval: When Small Embeddings and LLM Re-ranking Beat Bigger Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908387712696320 |
|---|---|
| author | Rao, Arjun Alipour, Hanieh Pendar, Nick |
| author_facet | Rao, Arjun Alipour, Hanieh Pendar, Nick |
| contents | This paper presents a comparison of embedding models in tri-modal hybrid retrieval for Retrieval-Augmented Generation (RAG) systems. We investigate the fusion of dense semantic, sparse lexical, and graph-based embeddings, focusing on the performance of the MiniLM-v6 and BGE-Large architectures. Contrary to conventional assumptions, our results show that the compact MiniLM-v6 outperforms the larger BGE-Large when integrated with LLM-based re-ranking within our tri-modal hybrid framework. Experiments conducted on the SciFact, FIQA, and NFCorpus datasets demonstrate significant improvements in retrieval quality with the MiniLM-v6 configuration. The performance difference is particularly pronounced in agentic re-ranking scenarios, indicating better alignment between MiniLM-v6's embedding space and LLM reasoning. Our findings suggest that embedding model selection for RAG systems should prioritize compatibility with multi-signal fusion and LLM alignment, rather than relying solely on larger models. This approach may reduce computational requirements while improving retrieval accuracy and efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_00049 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Rethinking Hybrid Retrieval: When Small Embeddings and LLM Re-ranking Beat Bigger Models Rao, Arjun Alipour, Hanieh Pendar, Nick Information Retrieval Artificial Intelligence This paper presents a comparison of embedding models in tri-modal hybrid retrieval for Retrieval-Augmented Generation (RAG) systems. We investigate the fusion of dense semantic, sparse lexical, and graph-based embeddings, focusing on the performance of the MiniLM-v6 and BGE-Large architectures. Contrary to conventional assumptions, our results show that the compact MiniLM-v6 outperforms the larger BGE-Large when integrated with LLM-based re-ranking within our tri-modal hybrid framework. Experiments conducted on the SciFact, FIQA, and NFCorpus datasets demonstrate significant improvements in retrieval quality with the MiniLM-v6 configuration. The performance difference is particularly pronounced in agentic re-ranking scenarios, indicating better alignment between MiniLM-v6's embedding space and LLM reasoning. Our findings suggest that embedding model selection for RAG systems should prioritize compatibility with multi-signal fusion and LLM alignment, rather than relying solely on larger models. This approach may reduce computational requirements while improving retrieval accuracy and efficiency. |
| title | Rethinking Hybrid Retrieval: When Small Embeddings and LLM Re-ranking Beat Bigger Models |
| topic | Information Retrieval Artificial Intelligence |
| url | https://arxiv.org/abs/2506.00049 |