Search papers, labs, and topics across Lattice.
3
0
6
4
PP-OCRv6 outperforms billion-scale VLMs on OCR tasks with a fraction of the parameters, achieving state-of-the-art accuracy and speed.
Targeted optimization in underperforming regions boosts document parsing accuracy to a record 96.33%, setting a new benchmark in the field.
Multi-task AV-LLMs can actually *improve* performance over single-task models, if you carefully design the training data and explicitly model inter-task relationships to avoid negative transfer.
