Camml: context-aware multimodal learner for large models

[免费ChatGPT] [DeepSeek可用网址] [PDF] arxiv.org

Cadvlm: Bridging language and vision in the generation of parametric cad sketches

S Wu, AH Khasahmadi, M Katz, PK Jayaraman… - … on Computer Vision, 2024 - Springer
Abstract Parametric Computer-Aided Design (CAD) is central to contemporary mechanical
design. However, it encounters challenges in achieving precise parametric sketch modeling …
Išsaugoti Cituoti Cituoja 2 Susiję straipsniai Visos 7 versijos
[免费ChatGPT] [DeepSeek可用网址] [PDF] arxiv.org

Memory Helps, but Confabulation Misleads: Understanding Streaming Events in Videos with MLLMs

G Zhang, M Ding, T Liu, Y Zhang, V Tresp - arxiv preprint arxiv …, 2025 - arxiv.org
Multimodal large language models (MLLMs) have demonstrated strong performance in
understanding videos holistically, yet their ability to process streaming videos-videos are …
Išsaugoti Cituoti Susiję straipsniai HTML kopija