Enhancing Visual Token Representations for Video Large Language Models via Training-Free Spatial-Temporal Pooling and Gridding

Documenti analoghi

• ST-SimDiff: Balancing Spatiotemporal Similarity and Difference for Efficient Video Understanding with MLLMs
  di: Luo, Bingjun, et al.
  Pubblicazione: (2026)
• ST$^3$: Accelerating Multimodal Large Language Model by Spatial-Temporal Visual Token Trimming
  di: Zhuang, Jiedong, et al.
  Pubblicazione: (2024)
• ForestPrune: High-ratio Visual Token Compression for Video Multimodal Large Language Models via Spatial-Temporal Forest Modeling
  di: Ju, Shaobo, et al.
  Pubblicazione: (2026)
• TS-LLaVA: Constructing Visual Tokens through Thumbnail-and-Sampling for Training-Free Video Large Language Models
  di: Qu, Tingyu, et al.
  Pubblicazione: (2024)
• EtC: Temporal Boundary Expand then Clarify for Weakly Supervised Video Grounding with Multimodal Large Language Model
  di: Li, Guozhang, et al.
  Pubblicazione: (2023)