Fréchet Distance optimized as FD-loss in representation space by decoupling population size from batch size improves generator quality, enables one-step generation from multi-step models, and motivates a multi-representation metric FDr^k.
Scaling laws for reward model overoptimization
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
CDM amortizes SMC inference for reward-tilted discrete diffusion by training a parameterized twist function on contrastive samples with closed-form kernels.
TMPO uses Softmax Trajectory Balance to match policy probabilities over multiple trajectories to a Boltzmann reward distribution, improving diversity by 9.1% in diffusion alignment tasks.
citing papers explorer
-
Representation Fr\'echet Loss for Visual Generation
Fréchet Distance optimized as FD-loss in representation space by decoupling population size from batch size improves generator quality, enables one-step generation from multi-step models, and motivates a multi-representation metric FDr^k.
-
Contrastive Distribution Matching for Amortized Sequential Monte Carlo in Discrete Diffusion
CDM amortizes SMC inference for reward-tilted discrete diffusion by training a parameterized twist function on contrastive samples with closed-form kernels.
-
TMPO: Trajectory Matching Policy Optimization for Diverse and Efficient Diffusion Alignment
TMPO uses Softmax Trajectory Balance to match policy probabilities over multiple trajectories to a Boltzmann reward distribution, improving diversity by 9.1% in diffusion alignment tasks.