FlashSampling: Fast and Memory-Efficient Exact Sampling

arXiv:2603.15854 · cs.LG, cs.AI, cs.CL · Submitted 2026-03-16 · Read on arXiv

cs.LG, cs.AI, cs.CL

Submitted: 2026-03-16

Updated: 2026-09-25

Comments: Project Page: https://github.com/FlashSampling/FlashSampling

Code: https://github.com/FlashSampling/FlashSampling

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Decode — 15× in this paper · explained in 2 episodes
• Core — 6× in this paper · explained in 2 episodes
• Ablation — 3× in this paper · explained in 1 episode
• Benchmark — 3× in this paper · explained in 2 episodes
• Time Per Output Token — 3× in this paper
• Adam — 2× in this paper · explained in 2 episodes
• Benchmarking — 2× in this paper · explained in 3 episodes
• Accuracy — 1× in this paper · explained in 1 episode
• Fidelity — 1× in this paper · explained in 2 episodes
• LLM judge — 1× in this paper · explained in 2 episodes
• Sparsity — 1× in this paper · explained in 4 episodes
• Transformer — 1× in this paper · explained in 2 episodes
• Verification — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Entropy-Aligned Decoding of LMs for Better Writing and Reasoning
• Robust benchmarking in noisy environments
• Training Verifiers to Solve Math Word Problems
• From Projection to Prediction: Beyond Logits for Scalable Language Models
• The Llama 3 Herd of Models
• The Curious Case of Neural Text Degeneration
• Approximate Top-$k$ for Increased Parallelism
• Online normalizer calculation for softmax
• FlashFormer: Whole-Model Kernels for Efficient Low-Batch Inference
• Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
• Qwen3 Technical Report
• Deep Kernel Fusion for Transformers

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