Transformer-based In-Context Score Operator Learning for Diffusion Models

Key points

• Diffusion models typically learn the generative dynamics of a single target distribution or a family of distributions indexed by conditioning variables fixed during training.1
• Our method learns an in-context score operator that maps the context samples, a noisy query point, and the diffusion time to the task-dependent score, enabling reverse-diffusion sampling on unseen tasks without retraining or parameter updates.1
• Its empirical Nadaraya--Watson approximation can be represented exactly by a model in our constructed transformer class, providing a principled foundation for the proposed query-dependent attention architecture.1
• The results show that the proposed method performs well on relatively simple task families, while its performance degrades as task complexity increases, thereby illustrating both the promise and the limitations of in-context learning for diffusion models.1

Sources (1)

• [1]Transformer-based In-Context Score Operator Learning for Diffusion Models
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 02:53 PM

  We introduce an in-context solution operator learning framework for diffusion models, in which each target distribution is treated as a task drawn from a meta-distribution and is specified at inference time by an empirical context set.

  Diffusion models typically learn the generative dynamics of a single target distribution or a family of distributions indexed by conditioning variables fixed during training.

Extractive summary: sentences quoted from the sources.