Text Encoder
المرمِّز النصي
مكوِّن في خط الأنابيب يُحوّل النص المُدخَل إلى تمثيلات رقمية (تضمينات) يمكن لنماذج أخرى استخدامها. في Imagen، المرمِّز النصي هو T5-XXL المُجمَّد.
A component in the pipeline that converts input text into numerical representations (embeddings) that other models can use. In Imagen, the text encoder is the frozen T5-XXL.
Also translated asمرمّز النصوص
First appears in this corpus in: ALIGN: Scaling Up Visual and Vision-Language Representation Learning with Noisy Text Supervision (2021)
Appears in these papers
- ALIGN: Scaling Up Visual and Vision-Language Representation Learning with Noisy Text Supervision2021in the sky ✦
- GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models2021in the sky ✦
- Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding2022in the sky ✦
- Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding2022in the sky ✦
- Do As I Can, Not As I Say: Grounding Language in Robotic Affordances2022in the sky ✦