🛰️ Daily AI Frontier
‹ back to 2026-09-09

Transformers as In-Context Samplers: From Closed-Form Diffusion to Estimation-Free Sampling

Research LLMs & Foundation Models

Ranking

Overall 78
Content 95
Popularity 37

Observed public metrics from 1 member.

Merged summary

TL;DR - This paper proves that frozen transformers can use prompt examples to simulate iterative generative samplers without parameter updates or explicit density estimation. It connects transformer components and layer-wise representation geometry to diffusion and energy-based sampling.

  • Softmax attention computes responsibility weights and weighted empirical averages, while feedforward layers implement Euler updates.
  • The constructions realize closed-form and smoothed closed-form diffusion samplers from in-context samples.
  • In semantic-topic experiments, normalized hidden states move toward a uniform spherical reference in intermediate layers, then recover topic-dependent structure near the output.
  • An interacting-particle energy follows the same U-shaped pattern across layers, which the authors reproduce theoretically with an approximate energy-based sampler.

Sources (1)

Transformers as In-Context Samplers: From Closed-Form Diffusion to Estimation-Free Sampling

arXiv cs.LG Arman Adibi, Alireza Jafari, Mohammad Ghavamzadeh, Hadi Daneshmand 2026-09-08 arXiv:2609.08981
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-19 14:16:08.528835 UTC

TL;DR - This paper proves that frozen transformers can use prompt examples to simulate iterative generative samplers without parameter updates or explicit density estimation. It connects transformer components and layer-wise representation geometry to diffusion and energy-based sampling.

  • Softmax attention computes responsibility weights and weighted empirical averages, while feedforward layers implement Euler updates.
  • The constructions realize closed-form and smoothed closed-form diffusion samplers from in-context samples.
  • In semantic-topic experiments, normalized hidden states move toward a uniform spherical reference in intermediate layers, then recover topic-dependent structure near the output.
  • An interacting-particle energy follows the same U-shaped pattern across layers, which the authors reproduce theoretically with an approximate energy-based sampler.
item →