LightSTAR achieves state-of-the-art accuracy in visual document retrieval by decomposing the task into LLM-free high-recall candidate selection and vision-adaptive semantic refinement on candidates, cutting end-to-end latency several-fold.
In: Computer Vision – ECCV 2022
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Introduces a non-causal attention refinement module to remove order dependence from cell representations in autoregressive table recognition models.
citing papers explorer
-
LightSTAR: Efficient Visual Document Retrieval via Lightweight Selection with Vision-Adaptive Refinement
LightSTAR achieves state-of-the-art accuracy in visual document retrieval by decomposing the task into LLM-free high-recall candidate selection and vision-adaptive semantic refinement on candidates, cutting end-to-end latency several-fold.
-
Revisiting Structural Dependency in Autoregressive Multi-Task Table Recognition via Order-Independent Cell-Level Representations
Introduces a non-causal attention refinement module to remove order dependence from cell representations in autoregressive table recognition models.