SynthDocQA
7.3
A platform for generating and evaluating visual document understanding models, leveraging synthetic datasets like SynthDocBench to systematically assess VLM performance in long-context scenarios, supporting more granular analysis and identifying areas for improvement.
200h
mvp estimate
7.3
viability grade
0
views
technology stack
Python
Medium
PostgreSQL
inspired by
SynthDocBench: A New Benchmark for Long-Context Visual Document Understanding Reveals VLM Weaknesses