To MRL or not to MRL: Text Embeddings are Robust to Truncation Without Matryoshka Learning, Except In Heavy Truncation Scenarios 文章

ArXiv CS.CL2026-05-29NEWSen作者: Sotaro Takeshita, Yurina Takeshita, Simone Paolo Ponzetto, Daniel Ruffinelli
查看原文 →
关系图谱

To MRL or not to MRL: Text Embeddings are Robust to Truncation Without Matryoshka Learning, Except In Heavy Truncation Scenarios · 相关技术

相关技术

Matryoshka Representation Learning