Guardado en:
Detalles Bibliográficos
Autores principales: Gill, Waris, Cechmanek, Justin, Hutcherson, Tyler, Rajamohan, Srijith, Agarwal, Jen, Gulzar, Muhammad Ali, Singh, Manvinder, Dion, Benoit
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2504.02268
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910902548168704
author Gill, Waris
Cechmanek, Justin
Hutcherson, Tyler
Rajamohan, Srijith
Agarwal, Jen
Gulzar, Muhammad Ali
Singh, Manvinder
Dion, Benoit
author_facet Gill, Waris
Cechmanek, Justin
Hutcherson, Tyler
Rajamohan, Srijith
Agarwal, Jen
Gulzar, Muhammad Ali
Singh, Manvinder
Dion, Benoit
contents This report investigates enhancing semantic caching effectiveness by employing specialized, fine-tuned embedding models. Semantic caching relies on embedding similarity rather than exact key matching, presenting unique challenges in balancing precision, query latency, and computational efficiency. We propose leveraging smaller, domain-specific embedding models, fine-tuned with targeted real-world and synthetically generated datasets. Our empirical evaluations demonstrate that compact embedding models fine-tuned for just one epoch on specialized datasets significantly surpass both state-of-the-art open-source and proprietary alternatives in precision and recall. Moreover, we introduce a novel synthetic data generation pipeline for the semantic cache that mitigates the challenge of limited domain-specific annotated data, further boosting embedding performance. Our approach effectively balances computational overhead and accuracy, establishing a viable and efficient strategy for practical semantic caching implementations.
format Preprint
id arxiv_https___arxiv_org_abs_2504_02268
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Advancing Semantic Caching for LLMs with Domain-Specific Embeddings and Synthetic Data
Gill, Waris
Cechmanek, Justin
Hutcherson, Tyler
Rajamohan, Srijith
Agarwal, Jen
Gulzar, Muhammad Ali
Singh, Manvinder
Dion, Benoit
Machine Learning
Computation and Language
This report investigates enhancing semantic caching effectiveness by employing specialized, fine-tuned embedding models. Semantic caching relies on embedding similarity rather than exact key matching, presenting unique challenges in balancing precision, query latency, and computational efficiency. We propose leveraging smaller, domain-specific embedding models, fine-tuned with targeted real-world and synthetically generated datasets. Our empirical evaluations demonstrate that compact embedding models fine-tuned for just one epoch on specialized datasets significantly surpass both state-of-the-art open-source and proprietary alternatives in precision and recall. Moreover, we introduce a novel synthetic data generation pipeline for the semantic cache that mitigates the challenge of limited domain-specific annotated data, further boosting embedding performance. Our approach effectively balances computational overhead and accuracy, establishing a viable and efficient strategy for practical semantic caching implementations.
title Advancing Semantic Caching for LLMs with Domain-Specific Embeddings and Synthetic Data
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2504.02268