BOLT: Boost Large Vision-Language Model Without Training for Long-form Video Understanding

Documenti analoghi

• End-to-End Temporal Action Detection with 1B Parameters Across 1000 Frames
  di: Liu, Shuming, et al.
  Pubblicazione: (2023)
• TimeLoc: A Unified End-to-End Framework for Precise Timestamp Localization in Long Videos
  di: Zhang, Chen-Lin, et al.
  Pubblicazione: (2025)
• $β$-CLIP: Text-Conditioned Contrastive Learning for Multi-Granular Vision-Language Alignment
  di: Zohra, Fatimah, et al.
  Pubblicazione: (2025)
• Video Self-Stitching Graph Network for Temporal Action Localization
  di: Zhao, Chen, et al.
  Pubblicazione: (2020)
• Learning Compact Video Representations for Efficient Long-form Video Understanding in Large Multimodal Models
  di: Chen, Yuxiao, et al.
  Pubblicazione: (2026)