# The output files are saved to `script_results` directory. # Script Results will contain a sub-directory for the date (`log_date_folder`). # The Log Date Folder will have sub-directories per Title-ID ...
We present ReconVLA, an implicit grounding paradigm for Vision-Language-Action models that reconstructs gaze regions to focus visual attention, achieving precise manipulation and strong generalization ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results