Hello Aparna,
Greetings and Welcome to Microsoft Q&A! Thanks for posting the question.
I understand that you're getting an error from Azure OpenAI's content filter when sending a specific Japanese text to the GPT-4 model.
I attempted to reproduce the issue in my environment, and it is working correctly for me. Here is the screenshot,
The issue could be due to environment and region differences, as Azure regions may have varying content moderation rules. Testing in another region might help identify discrepancies.
API configuration differences in model versions or deployment settings could cause inconsistencies.
Content filtering settings in the your environment may trigger the issue, and reviewing content filtering logs can provide clarity.
Also, consider this,
- Content validation to ensure the input adheres to expected standards and doesn't trigger unintended filters.
- Leverage content moderation tools to identify and resolve potentially problematic content before submission to the API.
- Consider using alternate APIs for content preprocessing, such as translation or simplification services, to refine the input without altering its meaning.
- Implementing segmentation and chunking can help break down large or complex inputs into manageable parts, reducing the likelihood of triggering content filters and ensuring smoother processing.
Kindly refer this Content filtering.
I Hope this helps. Do let me know if you have any further queries.
If the response helped, please do click Accept Answer
and Yes
for was this answer helpful.
Thank you!