method · first seen 26 Oct 2026
value gradient guidance
Guiding diffusion sampling using the gradient of a value function estimating expected reward given a noisy state.
Mentions per week
PapersCodeDiscussion
Earliest sources
- VaLoP: Value Gradient Lookahead and Pullback for Reward-Guided DiffusionPapers · 26 Oct 2026 · Kwanyoung Kim
Latest sources
- VaLoP: Value Gradient Lookahead and Pullback for Reward-Guided DiffusionPapers · 26 Oct 2026 · Kwanyoung Kim