Skip to main content
Skip table of contents

Resetting the Contextual AI & Purging or Preserving Memories

Contextual AI suggestions can degrade over time if poor Translation Memory (TM) entries are uploaded to your Data Sources. To help you manage the quality of Contextual AI models, the Lilt platform allows you to do the following:

  • Reset a Contextual AI model.
  • Purge a Data Source.
  • Preserve a Data Source.

See the TM Size and Content Guidelines article for guidelines on what TM data is best suited for Lilt.

Resetting the Contextual AI model

If you feel the Contextual AI quality for a given Data Source has degraded over time, you can Reset the model by following these steps:

  1. Open the Data > Sources tab.
  2. Open the desired Data Source by clicking on the name or clicking the Edit button.
  3. Open the Settings page.
  4. Under the Reset MT section, click the Reset button to open a popup where you will need to confirm you want to reset the model.

Reset the Contextual AI to perform the following:

  • reset the Lilt Translate automatic suggestions to a clean slate
  • re-train the Contextual AI model on all confirmed segments in the Data Source

Note: when resetting the Contextual AI, your TB and TM entries will remain untouched.

Purging memory and termbase entries

You can restore your Data Source to a clean slate by purging the memory and termbase entries:

  1. Open the Data > Sources tab.
  2. Open the desired Data Source by clicking on the Translation Memory name or clicking the Edit button.
  3. Open the Settings page.
  4. Under the Purge memory & termbase section, click the Purge memory and termbase button to open a popup where you will need to confirm you want to purge the Data Source.

Purge the Translation Memory & Termbase to perform the following:

  • remove all the custom TB and TM entries from the Data Source
  • reset the Lilt Translate automatic suggestions to a clean state; does not retrain on the existing TM data

Preserving Data Source by disabling in-context learning

By default, Data Source adapt to both uploaded TMX files and the linguistic activity of translators and reviewers in Lilt Translate. If at any point you need to stop a Data Source from adapting any further, you can do so by disabling the in-context learning option within the Data Source settings.

Disabling in-context learning when creating a new Data Source

During the creation of a new Data Source, toggle off the option for in-context learning and then continue with creating the Data Source.

Disabling in-context learning on an existing Data Source

  1. Open the Data > Sources tab.
  2. Open the desired Data Source by clicking on the name or clicking the Edit button.
  3. Open the Settings page.
  4. Toggle in-context learning off. This can be enabled/disabled at any time.

JavaScript errors detected

Please note, these errors can depend on your browser setup.

If this problem persists, please contact our support.