Created on December 17, 2025
2025
[ArXiv] One first author paper, the first vision-text compression benchmark: VTCBench.