TextDS uses a data-efficient dual-encoder with SWLoRA and CSF to achieve competitive scene text detection robustness under distribution shifts and adverse conditions using 4.9M trainable parameters.
arXiv preprint arXiv:2510.16442 (2025) 14
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
TextDS: Parameter-Efficient Representation Alignment for Scene Text Detection under Distribution Shifts
TextDS uses a data-efficient dual-encoder with SWLoRA and CSF to achieve competitive scene text detection robustness under distribution shifts and adverse conditions using 4.9M trainable parameters.