Kind request to include "More Thought, Less Accuracy? On the Dual Nature of Reasoning in Vision-Language Models"

[xytian1008]

Dear Team,

Thank you for curating such a valuable collection of recent vision-language models works. We’d be grateful if you could consider adding our new paper:

“More Thought, Less Accuracy? On the Dual Nature of Reasoning in Vision-Language Models.”

arXiv: https://www.arxiv.org/abs/2509.25848
Project: https://xytian1008.github.io/VAPO/
GitHub: https://github.com/xytian1008/VAPO
Many thanks!

Best regards,
Xinyu