An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models

Documenti analoghi

• A Spark of Vision-Language Intelligence: 2-Dimensional Autoregressive Transformer for Efficient Finegrained Image Generation
  di: Chen, Liang, et al.
  Pubblicazione: (2024)
• Multimodal Representation Alignment for Image Generation: Text-Image Interleaved Control Is Easier Than You Think
  di: Chen, Liang, et al.
  Pubblicazione: (2025)
• Looking Beyond Text: Reducing Language bias in Large Vision-Language Models via Multimodal Dual-Attention and Soft-Image Guidance
  di: Zhao, Haozhe, et al.
  Pubblicazione: (2024)
• VLCache: Computing 2% Vision Tokens and Reusing 98% for Vision-Language Inference
  di: Qin, Shengling, et al.
  Pubblicazione: (2025)
• Turbo: Informativity-Driven Acceleration Plug-In for Vision-Language Large Models
  di: Ju, Chen, et al.
  Pubblicazione: (2024)