# October 2023 second release

## **📢 Announcement: Data retention changes are live**

The following data was not included in the scope of "data retention" before but will now be automatically removed if it is older than the configured data retention period:\
\
1\. **Event logs.** The user actions in the event logs will be automatically removed after the configured data retention period to comply with privacy regulations.\
2\. **Training documents with status 'Input required'.** We have noticed many projects contain outdated documents that were sent automatically from production to training long ago. As a best practice, it's better to add recent training data compared to old, and by reducing the number of Input Required documents, the QA tasks for improving models become more manageable.<br>

Furthermore, for new projects only, the **default data retention period will be 6 months** instead of 2 years.

{% content-ref url="/pages/d4iR0N426gZTZW3XMcvU" %}
[Data retention](/project/project-settings/data-retention.md)
{% endcontent-ref %}

<figure><img src="/files/3H5ZHjH1TuBprRZfYg1j" alt=""><figcaption></figcaption></figure>

## **🔮 Annotation-less models \[🔬 Experimental]**

The annotation-less models based on OpenAI GPTx functionality has been improved to also work for composites. You can now get predictions for your line items in your documents when using the annotation-less models!

<figure><img src="/files/PDEiDZsMdcsQYqltFiH7" alt=""><figcaption></figcaption></figure>

{% embed url="<https://docs.app.metamaze.eu/project/project-settings/general-settings#openai-gptx>" %}

<figure><img src="/files/3H5ZHjH1TuBprRZfYg1j" alt=""><figcaption></figcaption></figure>

## **🔎 Filter validation list improvement**

New more general filters have been added when filtering the validation queue based on "Parked by":

* Documents parked someone
* Document parked by no one

<figure><img src="/files/mZcIfFyzVfU9YRVSmQdV" alt=""><figcaption></figcaption></figure>

{% embed url="<https://docs.app.metamaze.eu/project/production/validation#parking-documents>" %}

<figure><img src="/files/3H5ZHjH1TuBprRZfYg1j" alt=""><figcaption></figcaption></figure>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs-old.app.metamaze.eu/release-notes/release-archive/release-archive-2023/october-2023-second-release.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.