Open Models Almanac
Methods

single mention · not yet filed

Soft prompting with learnable embodiment-specific data-source embeddings

Soft prompting with learnable embodiment-specific data-source embeddings is a mechanism that uses learnable prompt embeddings to explicitly absorb embodiment-specific variability.1 By associating each hardware configuration with learnable prompt embeddings, the model captures domain-specific knowledge to enable stable large-scale pretraining.2

How this definition was written

Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.

  1. we introduce a soft-prompt mechanism that explicitly absorbs embodiment-specific variability while preserving a shared backbone for general reasoning — X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model
  2. By associating each hardware configuration with learnable prompt embeddings, the model can flexibly capture domain-specific knowledge, thus enabling stable large-scale pretraining — X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model

Also called Soft Prompt.