Zhenhailong Wang
@zhenhailongW • 1,294 subscribers
Incoming AP Westlake University; Research scientist at Apodex; Ph.D. at UIUC, advised by Prof. Heng Ji. Intern at Tencent, MSRA, Salesforce, Amazon
Videos
Large multimodal models often lack precise low-level perception needed for high-level reasoning, even with simple vector graphics. We bridge this gap by proposing an intermediate symbolic representation that leverages LLMs for text-based reasoning. 🧵1/4
Zhenhailong Wang37,320 görüntüleme • 2 yıl önce
Daha fazla içerik yok.