Docling Technical Report at AAAI 25 Workshop on Open-Source AI for Mainstream Use #574
dolfim-ibm
announced in
Announcements
Replies: 1 comment
-
|
May be using RapidOCR instead of EasyOCR will get a better E2E performance result than Miner-U or Unstructured. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
We are happy to announce that the Docling Technical Report got accepted for the AAAI 25: Workshop on Open-Source AI for Mainstream Use.
The paper (attached to this discussion) explains the Docling architecture and provides reference numbers when running on CPU, L4 GPU, Macbook M3 Max. The numbers are also compared to other mainstream projects like Marker, MinerU, Unstructured.
The proper citation for it will be added soon.
AAAI_2025_workshop.pdf
Beta Was this translation helpful? Give feedback.
All reactions