Search papers, labs, and topics across Lattice.
This paper introduces LePaX, a novel framework for chest X-ray report generation that enables high-resolution perception without exceeding the fixed visual token budgets of pretrained vision encoders. By employing Learnable Spatial Resolution Allocation (LSRA) and Global-Regional Fusion (GRF), LePaX effectively localizes and enhances diagnostically relevant regions in CXR images, allowing for detailed analysis while maintaining a manageable token count. Experimental results show that LePaX significantly improves clinical and linguistic metrics on multiple benchmarks, achieving native-resolution perception with over 10 times fewer visual tokens than traditional methods.
High-resolution chest X-ray perception can be achieved without inflating token counts, leading to better diagnostic accuracy and report generation.
Despite rapid advances in chest X-ray (CXR) foundation models, most radiology report generation (RRG) systems still rely on heavily downsampled inputs (e.g., 256x256) due to the fixed visual token budgets of pretrained vision encoders, suppressing subtle yet clinically important cues present in native-resolution images. However, enabling high-resolution (high-res) perception remains challenging: naive tiling causes prohibitive token inflation, while global compression suppresses subtle lesions and degrades diagnostic fidelity. Inspired by radiologists'workflow, localizing suspicious regions before detailed high-res assessment. We propose Lesion-Aware High-Resolution Patch Discovery and Fusion for Chest X-ray Reporting (LePaX), the first RRG framework that enables efficient high-res CXR perception (up to 1920x1920) without increasing the vision-token count. LePaX formulates high-res perception as a constrained spatial resolution allocation problem under a fixed token budget and introduces two key components: Learnable Spatial Resolution Allocation (LSRA), which learns a spatial utility map that adaptively allocates limited high-res capacity to diagnostically relevant regions, enabling targeted extraction of high-res patches from native CXRs; and Global-Regional Fusion (GRF), which performs token-preserving region-to-global refinement by projecting high-resolution regional evidence back onto the global feature grid through spatially aligned resolution write-back, avoiding token inflation. Experiments on multiple CXR benchmarks demonstrate that LePaX consistently improves both clinical and linguistic metrics while enabling native-resolution CXR perception with over 10x fewer visual tokens than naive high-res tiling.
