Data Normalization

Data processing and normalization tasks evaluate structured output from messy prose and different structures. This capability is essential for catalogue and product pipelines where data needs to be extracted from unstructured text and formatted consistently. Models must understand the desired output format and extract relevant information accurately.

50
Input Tokens
0
Output Tokens
0
Duration
0
Cost
0
Context
Input
Analyze the following cooking instructions and return a structured response following these criteria:
1. Return a list of all ingredients present in the recipe even if they are optional (Only ingredients, not quantities)
2. Ingredients should only be mentioned once
3. Exclude the title
Expected output
{
  "ingredients": [
    "basmati rice",
    "bay leaf",
    "chicken",
    "cilantro",
    "cinnamon",
    "cloves",
    "garam masala",
    "ghee",
    "ginger-garlic paste",
    "green cardamom",
    "lemon",
    "milk",
    "mint",
    "oil",
    "onions",
    "red chili powder",
    "saffron",
    "salt",
    "turmeric",
    "water",
    "yogurt"
  ]
}
Model output
{
  "ingredients": [
    "basmati rice",
    "water",
    "bay leaf",
    "cloves",
    "green cardamom pods",
    "cinnamon stick",
    "salt",
    "chicken",
    "plain yogurt",
    "ginger-garlic paste",
    "turmeric",
    "red chili powder",
    "garam masala",
    "lemon",
    "oil",
    "onions",
    "cilantro",
    "mint",
    "milk",
    "saffron",
    "ghee"
  ]
}