Mistakes parsing data from table using LlamaParse and gpt4o #124

Open
opened 2026-02-16 00:16:56 -05:00 by yindo · 2 comments
Owner

Originally created by @xmanatsf on GitHub (May 29, 2024).

Trying to extract tabular data (table is embedded as an image) from a PDF file. While I've managed to extract some data, there are consistent errors when the table is located at the bottom of the PDF. Extractions from tables at the top of the document are accurate. I'm not sure what's causing this discrepancy.

Originally created by @xmanatsf on GitHub (May 29, 2024). Trying to extract tabular data (table is embedded as an image) from a PDF file. While I've managed to extract some data, there are consistent errors when the table is located at the bottom of the PDF. Extractions from tables at the top of the document are accurate. I'm not sure what's causing this discrepancy.
Author
Owner

@iiitmahesh commented on GitHub (May 29, 2024):

Same issue for me.

@iiitmahesh commented on GitHub (May 29, 2024): Same issue for me.
Author
Owner

@zheqiaochen commented on GitHub (Jan 17, 2025):

I have the same issue, it seems that the tables at the bottom could not be parsed

@zheqiaochen commented on GitHub (Jan 17, 2025): I have the same issue, it seems that the tables at the bottom could not be parsed
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: run-llama/llama_cloud_services#124