> ## Documentation Index
> Fetch the complete documentation index at: https://www.getmaxim.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Summarization

> Evaluates the quality of text summarization across multiple dimensions including content coverage, alignment with source, and information accuracy.

## Input

* **Required Inputs:**
  * **`input`**: Original text
  * **`output`**: The generated summary text to be evaluated

## Output

* **`Result`**: Value in the continuous range \[0, 1]
* **`Reasoning`**: Detailed explanation of assessment

The Summarization evaluator score is calculated using the following equation:

$$
\mathrm{Summarization} = \min(\text{Alignment Score}, \text{Coverage Score})
$$

To break it down:

* **Alignment Score**: Determines whether the summary contains hallucinated or contradictory information compared to the original text.
* **Coverage Score**: Determines whether the summary includes the necessary information from the original text.

## Interpretation

* **Higher scores**: Indicate better summarization quality
* **Lower scores**: Indicate missing information or inaccuracies