1 paper · 1 filter
Brandon Smock, Libin Liang, Max Sokolov +4
Large-scale document processing requires contextually aware table extraction (TE) that is both accurate and efficient. Yet current approaches require billions of parameters, hundre…