Researchers fine-tune 560M-parameter E5 embedder for Tajik passage retrieval
A new arXiv paper describes Saidzoda-Tajik-E5-Large, a 560M-parameter embedding model fine-tuned for Tajik question-answering passage retrieval, filling a gap in openly described models for the language.
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-11-03 05:00 · arXiv cs.CL
Saidzoda-Tajik-E5-Large: Fine-Tuning a 560M Embedder for Tajik Retrieval