Rank-Aware Speculative Sampling for Diffusion Draft Trees
cs.LG, stat.CO
Submitted: 2026-09-30
Updated: 2026-09-30
Sources
- Stochastic Interpolants: A Unifying Framework for Flows and Diffusions
- Accelerating Diffusion Sampling via Speculative Draft Trees
- Accelerating Large Language Model Decoding with Speculative Sampling
- Scalable Differentially Private Data Compression via Diffusion and Stochastic Codes
- SpecTr-GBV: Multi-Draft Block Verification Accelerating Speculative Decoding
- Multi-Marginal Couplings for Metropolis-Hastings
- Accelerating Speculative Diffusions via Block Verification
- Continuous Diffusion Scales Competitively with Discrete Diffusion for Language
- Dual-Representation Image Compression at Ultra-Low Bitrates via Explicit Semantics and Implicit Textures
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks