When and How Much to Imagine: Adaptive Test-Time Scaling with World Models for Visual Spatial Reasoning

Documenti analoghi

• SAFREE: Training-Free and Adaptive Guard for Safe Text-to-Image And Video Generation
  di: Yoon, Jaehong, et al.
  Pubblicazione: (2024)
• CREMA: Generalizable and Efficient Video-Language Reasoning via Multimodal Modular Fusion
  di: Yu, Shoubin, et al.
  Pubblicazione: (2024)
• RACCooN: A Versatile Instructional Video Editing Framework with Auto-Generated Narratives
  di: Yoon, Jaehong, et al.
  Pubblicazione: (2024)
• Video-RTS: Rethinking Reinforcement Learning and Test-Time Scaling for Efficient and Enhanced Video Reasoning
  di: Wang, Ziyang, et al.
  Pubblicazione: (2025)
• MEXA: Towards General Multimodal Reasoning with Dynamic Multi-Expert Aggregation
  di: Yu, Shoubin, et al.
  Pubblicazione: (2025)