Qianfan-VL: Domain-Enhanced Universal Vision-Language Models Paper • 2509.18189 • Published Sep 19, 2025 • 1