TeleOCR: A 1.2B Vision-Language Model for Structured Document Parsingtranslating…

Explore TeleOCR’s document-parsing benchmarks, distorted-page handling, table and formula extraction, limitations, and comparisons with other OCR VLMs.

Factuality

78/100

Ownership

Hacker Media House

Country

Unknown

Keywords#Document#Document Parsing#Model#Structured#Structured Document

Read Original ↗