Guardado en:
Detalles Bibliográficos
Autores principales: Huang, Junqin, Hu, Zhongjie, Jing, Zihao, Gao, Mengya, Wu, Yichao
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2405.06932
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909200056057856
author Huang, Junqin
Hu, Zhongjie
Jing, Zihao
Gao, Mengya
Wu, Yichao
author_facet Huang, Junqin
Hu, Zhongjie
Jing, Zihao
Gao, Mengya
Wu, Yichao
contents In this report, we introduce Piccolo2, an embedding model that surpasses other models in the comprehensive evaluation over 6 tasks on CMTEB benchmark, setting a new state-of-the-art. Piccolo2 primarily leverages an efficient multi-task hybrid loss training approach, effectively harnessing textual data and labels from diverse downstream tasks. In addition, Piccolo2 scales up the embedding dimension and uses MRL training to support more flexible vector dimensions. The latest information of piccolo models can be accessed via: https://huggingface.co/sensenova/
format Preprint
id arxiv_https___arxiv_org_abs_2405_06932
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Piccolo2: General Text Embedding with Multi-task Hybrid Loss Training
Huang, Junqin
Hu, Zhongjie
Jing, Zihao
Gao, Mengya
Wu, Yichao
Computation and Language
Artificial Intelligence
In this report, we introduce Piccolo2, an embedding model that surpasses other models in the comprehensive evaluation over 6 tasks on CMTEB benchmark, setting a new state-of-the-art. Piccolo2 primarily leverages an efficient multi-task hybrid loss training approach, effectively harnessing textual data and labels from diverse downstream tasks. In addition, Piccolo2 scales up the embedding dimension and uses MRL training to support more flexible vector dimensions. The latest information of piccolo models can be accessed via: https://huggingface.co/sensenova/
title Piccolo2: General Text Embedding with Multi-task Hybrid Loss Training
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2405.06932