PromptSync: Bridging Domain Gaps in Vision-Language Models through Class-Aware Prototype Alignment and Discrimination

Ähnliche Einträge

• RePoseDM: Recurrent Pose Alignment and Gradient Guidance for Pose Guided Image Synthesis
  von: Khandelwal, Anant
  Veröffentlicht: (2023)
• FloCoDe: Unbiased Dynamic Scene Graph Generation with Temporal Consistency and Correlation Debiasing
  von: Khandelwal, Anant
  Veröffentlicht: (2023)
• FlexiClip: Locality-Preserving Free-Form Character Animation
  von: Khandelwal, Anant
  Veröffentlicht: (2025)
• EmbodiedMidtrain: Bridging the Gap between Vision-Language Models and Vision-Language-Action Models via Mid-training
  von: Du, Yiyang, et al.
  Veröffentlicht: (2026)
• MSR-Align: Policy-Grounded Multimodal Alignment for Safety-Aware Reasoning in Vision-Language Models
  von: Xia, Yinan, et al.
  Veröffentlicht: (2025)