Piccolo2: General Text Embedding with Multi-task Hybrid Loss Training
This work addresses the need for improved text embeddings in natural language processing, representing an incremental advancement over previous models.
The authors tackled the problem of general text embedding by introducing Piccolo2, which achieved a new state-of-the-art on the CMTEB benchmark across 6 tasks, surpassing other models in comprehensive evaluation.
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/