LLM Digest

Story

arxiv_cs_lg · Oct 29, 2026 · paper

Source brief

Composing and Steering Diffusion Models with Generalized Feynman-Kac Correctors

arxiv.orgOct 29, 2026
original source linked

In brief

Diffusion models have become the leading paradigm in generative modeling, yet many applications require sampling from distributions that are not those of any single pretrained model, but are instead defined through combinations of several models, possibly tilted toward high value

Read the original at arxiv.org →Open in live feed Read that day’s brief

Composing and Steering Diffusion Models with Generalized Feynman-Kac Correctors

Earlier in this thread 4 items

Metacognitive Decoding: Efficient Inference for Diffusion Language Agents

Learning When to Commit: Schedule Distillation for Fast Diffusion Language Models

VaLoP: Value Gradient Lookahead and Pullback for Reward-Guided Diffusion

CANS: Capability-Preserving Neuron Steering for Large Language Models