Kuiz berkaitan panduan · Sederhana Tahap

Penalaan Halus lwn RAG lwn Kuiz Mendorong

Menguji cara untuk mendiagnosis pengetahuan, format, gaya dan jurang kemahiran, apabila RAG mengatasi penalaan halus, dan sebab set penilaian mendorong keputusan.

Laluan panduan yang berkaitanPenalaan Halus Vs RAG Vs Prompting
soalan 1 daripada 8

Model menjawab soalan tentang harga syarikat anda secara salah kerana harga berubah setiap suku tahun. Pendekatan manakah yang paling sesuai dengan jurang ini?

Pratonton setiap soalan8 soalan

Jawapan dan penjelasan muncul semasa anda mengambil kuiz.

  1. Model menjawab soalan tentang harga syarikat anda secara salah kerana harga berubah setiap suku tahun. Pendekatan manakah yang paling sesuai dengan jurang ini?
    • Fine-tune the model on last quarter's price list
    • Use RAG to retrieve the current price documents at answer time
    • Lower the model's temperature
    • Add more examples of polite answers to the prompt
  2. Menurut panduan, apakah biasanya tuas pertama untuk mencuba masalah format dan gaya?
    • Full fine-tuning
    • Building a vector database
    • Prompting with clear instructions and examples
    • Switching to a larger model
  3. Mengapakah penalaan halus digambarkan sebagai cara yang tidak boleh dipercayai untuk menambah pengetahuan fakta?
    • It cannot change model weights
    • Ia hanya berfungsi pada model imej
    • It is hard to update and can increase confident errors about rarely seen material
    • It always deletes the model's existing knowledge
  4. Situasi manakah yang paling jelas menggemari penalaan halus?
    • You need a narrow task done at high volume and want a smaller, cheaper model to match a larger one
    • Your documents change every day
    • You want answers to cite source passages
    • You have not yet built any evaluation examples
  5. Dalam sistem RAG yang memberikan jawapan yang salah, apakah yang perlu anda semak sebelum menyalahkan model bahasa?
    • Tarikh latihan model
    • Whether the correct passage appears in the retrieved results
    • The number of GPU cores
    • The length of the system prompt only
  6. What does LoRA do in the context of fine-tuning?
    • Retrieves documents from a search index
    • Compresses prompts to save tokens
    • Trains small adapter matrices instead of updating all model weights
    • Removes the need for training data
  7. Which statement about combining the three methods is accurate according to the guide?
    • They are mutually exclusive and cannot be combined
    • Production systems often use a fine-tuned model reading retrieved context, driven by a good prompt
    • RAG replaces the need for any prompt
    • Fine-tuning must always come before prompting
  8. What should drive the choice between prompting, RAG and fine-tuning?
    • Whichever method is newest
    • A representative evaluation set measuring each option
    • The vendor's default setting
    • The size of the training budget alone