Cut to the Chase: Training-free Multimodal Summarization via Chain-of-Events

Documenti analoghi

• Knowledge Completes the Vision: A Multimodal Entity-aware Retrieval-Augmented Generation Framework for News Image Captioning
  di: You, Xiaoxing, et al.
  Pubblicazione: (2025)
• Flexiffusion: Training-Free Segment-Wise Neural Architecture Search for Efficient Diffusion Models
  di: Huang, Hongtao, et al.
  Pubblicazione: (2025)
• Measuring Social Bias in Vision-Language Models with Face-Only Counterfactuals from Real Photos
  di: Chen, Haodong, et al.
  Pubblicazione: (2026)
• Training-free Composite Scene Generation for Layout-to-Image Synthesis
  di: Liu, Jiaqi, et al.
  Pubblicazione: (2024)
• Training-free Guidance in Text-to-Video Generation via Multimodal Planning and Structured Noise Initialization
  di: Li, Jialu, et al.
  Pubblicazione: (2025)