TTF-VLA: Temporal Token Fusion via Pixel-Attention Integration for Vision-Language-Action Models

Documenti analoghi

• MSNav: Zero-Shot Vision-and-Language Navigation with Dynamic Memory and LLM Spatial Reasoning
  di: Liu, Chenghao, et al.
  Pubblicazione: (2025)
• VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching
  di: Xu, Siyu, et al.
  Pubblicazione: (2025)
• PixelVLA: Advancing Pixel-level Understanding in Vision-Language-Action Model
  di: Liang, Wenqi, et al.
  Pubblicazione: (2025)
• VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers
  di: Wang, Yating, et al.
  Pubblicazione: (2025)
• PD-VLA: Accelerating Vision-Language-Action Model Integrated with Action Chunking via Parallel Decoding
  di: Song, Wenxuan, et al.
  Pubblicazione: (2025)