Multi-Line Invoice Item Processing produces unexpected results (Invoice Model - V4)

[rudiv]

This isn't really documentation related, but I'm not sure where to post issues with the service.

With V4's GA around the corner, I'm not exactly sure where to post issues with the preview model that are coming up.

We receive invoices where the Date is important on a per line basis, and we've noticed that multi-line dates are not parsed correctly.

Here's a sample of the dates in question, where the "Date" comes out as 2024 with relatively high confidence:

"Date": {
                    "type": "date",
                    "content": "2024",
                    "boundingRegions": [
                      {
                        "pageNumber": 1,
                        "polygon": [
                          3.9038,
                          4.5678,
                          4.137,
                          4.5678,
                          4.137,
                          4.6793,
                          3.8987,
                          4.6793
                        ]
                      }
                    ],
                    "confidence": 0.938,
                    "spans": [
                      {
                        "offset": 445,
                        "length": 4
                      }
                    ]
                  },

This is however read correctly (albeit with the newline token) in the Table view:

{
            "rowIndex": 1,
            "columnIndex": 1,
            "content": "Mon, 20-May-\n2024",
            "boundingRegions": [
              {
                "pageNumber": 1,
                "polygon": [
                  3.5633,
                  4.3917,
                  4.4761,
                  4.3917,
                  4.4761,
                  4.7202,
                  3.5633,
                  4.7202
                ]
              }
            ],
            "spans": [
              {
                "offset": 395,
                "length": 12
              },
              {
                "offset": 445,
                "length": 4
              }
            ]
          },

Shots from Intelligence Studio.

Invoice Row Date

Table "Service Date" (with inconsistent spacing / newline token)

I can share the invoice privately, or if I need to post / email / report this elsewhere please let me know.

---

Document Details

⚠ Do not edit this section. It is required for learn.microsoft.com ➟ GitHub issue linking.

• ID: 5f4dded6-e7ed-babc-4caf-a4e13bba97f7
• Version Independent ID: 7c118774-c2b1-86e1-0db6-aad156581c0c
• Content: Invoice data extraction – Document Intelligence (formerly Form Recognizer) - Azure AI services
• Content Source: articles/ai-services/document-intelligence/concept-invoice.md
• Service: azure-ai-document-intelligence
• GitHub Login: @laujan
• Microsoft Alias: lajanuar

[PesalaPavan]

@rudiv Thanks for your feedback! We will investigate and update as appropriate.

[Naveenommi-MSFT]

Hello @rudiv Thank you for your feedback!

Since this issue isn't directly related to improving our docs, and to gain a better understanding of your issue, I'd recommend working closer with our support team via an Azure support request. Or you can leverage our Q&A forum by posting your issue there so our community, and MVPs can further assist you in troubleshooting this issue or finding potential workarounds. Teams Q&A forum for technical questions about the configuration and administration of Microsoft Teams on Windows. Microsoft Teams Community forum.

[Naveenommi-MSFT]

@rudiv We are going to close this thread, if there are any further questions regarding the documentation, please tag me in your reply and we will be happy to continue the conversation.

Please-close