Post
92
Your documents have headings. Does your RAG chunker use them? π
I put together OutlineLM π€β a couple of models to help prepare meaningful chunks for RAG from Russian and English documents.
The focus is on headings that introduce a new semantic block, providing useful boundaries for splitting your documentsπππ
Use the detected headings as chunk boundaries, apply size limits where needed, and pass the chunks to your retrieval pipeline.
π Models & examples:
β’ DiTy/OutlineLM-Gemma4-E2B-v1
β’ DiTy/OutlineLM-Gemma4-E4B-v1
Give them a try β Iβd love to hear how they handle your documents
I put together OutlineLM π€β a couple of models to help prepare meaningful chunks for RAG from Russian and English documents.
The focus is on headings that introduce a new semantic block, providing useful boundaries for splitting your documentsπππ
Use the detected headings as chunk boundaries, apply size limits where needed, and pass the chunks to your retrieval pipeline.
π Models & examples:
β’ DiTy/OutlineLM-Gemma4-E2B-v1
β’ DiTy/OutlineLM-Gemma4-E4B-v1
Give them a try β Iβd love to hear how they handle your documents