Guide-yakabatana mibvunzo · Pakati Nhanho
Multimodal RAG Quiz
Edza zvaunoziva nezve multimodal RAG: zvinyorwa zvinoonekwa mumavara, akagovaniswa mufananidzo-mavara embeddings, uye ColPali-maitiro peji-screenshot kudzoreredza.
Mubvunzo 1 ye 8
Chii chinowanzoitika chisina kunaka kana pombi-chete yeRAG pombi inogadzira PDF izere nematafura nemachati?
Tarisa mubvunzo wese8 mibvunzo
Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.
- Chii chinowanzoitika chisina kunaka kana pombi-chete yeRAG pombi inogadzira PDF izere nematafura nemachati?
- The PDF becomes too small to index
- Tables become jumbled and chart values or diagrams are lost
- The embeddings become images
- The model refuses to answer
- Ndeipi kushaya simba kukuru kwecaption-everything-mu-text approach?
- It cannot use existing text embeddings
- It requires a vector database for images
- It only works for scanned forms
- Anything the captioner fails to describe is lost to retrieval
- Chii chinoita CLIP-maitiro modhi inoita kuti igoneke mune multimodal RAG?
- Embedding images and text in a shared vector space so text queries can retrieve images
- Reading dense small text on scanned pages with high accuracy
- Compressing PDFs before storage
- Translating captions into other languages
- Chii chinosiyanisa maitiro eColPali?
- It converts every page to Markdown with OCR first
- It only indexes image captions
- It embeds page screenshots with a vision-language model and skips OCR and layout parsing
- It stores one vector per document
- Mukunonoka kusangana, mubvunzo unopihwa sei pane peji?
- One query vector is compared with one page vector
- Pages are ranked by file size
- Keyword counts are compared
- Many query-token vectors are compared with many image-patch vectors
- Why are ColPali-style indexes much larger than text-chunk indexes?
- They store the original PDF files inside each vector
- Each page gets roughly one vector per image patch instead of a single vector
- They duplicate every page in several languages
- They require storing the full model weights per page
- Why pass the page image, not only extracted text, to the generating model?
- Questions about chart values or visual layout need the model to see the visual itself
- Images are always cheaper than text
- Text cannot be cited
- Generators cannot read text
- Which statement about multimodal RAG in practice is accurate?
- It always replaces text pipelines entirely
- It only works on photographs
- Hybrid systems combining parsed text, table extraction and page-image retrieval are common
- It requires no evaluation