SonicSampler: Unified Tile-Aware Kernels for LLM Sampling and Speculative Verification

• URL: http://arxiv.org/abs/2607.20475v1
• Date: Sun, 24 May 2026 23:01:50 GMT
• Title: SonicSampler: Unified Tile-Aware Kernels for LLM Sampling and Speculative Verification
• Abstract summary: We present $textbfSonicSampler$, a unified suite of tile-aware Triton kernels.<n>Our kernels support dynamic per-request sampling behaviors, including grammar-constrained decoding.<n>Across speculative decoding, SonicSampler achieves up to $textbf16x speedup over state-of-the-art baselines.
• Score: 27.358689380362964
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Sampling in LLM inference comprises a combinatorial set of logit processing, token selection, and verification operations for speculative decoding. However, existing implementations either accelerate only subsets of this pipeline, rely on multiple kernel launches, or assume homogeneous sampling behavior across a batch, limiting support for dynamic serving workloads and preventing efficient CUDA Graph execution. We present $\textbf{SonicSampler}$, a unified suite of tile-aware Triton kernels that vertically fuses the complete sampling pipeline into a fixed, workload-aware execution model. Our kernels support dynamic per-request sampling behaviors, including grammar-constrained decoding, repetition, frequency and presence penalties, logit bias, temperature scaling, top-$k$ / top-$p$ / min-$p$ filtering, and speculative verification - within a single batched kernel while remaining fully CUDA Graph-compatible. Central to our approach is a novel hierarchical two-stage top-$k$ algorithm that achieves up to $\textbf{10x speedup}$ over competitive baselines and exploits the low-entropy structure of LLM outputs to enable efficient selection over large vocabularies. Across heterogeneous speculative decoding workloads, SonicSampler achieves up to $\textbf{16x speedup}$ over state-of-the-art baselines while preserving flexible batched execution.

Related papers

• FluxBin: Flexible LUT-based Ultra-low-bit LLM Inference by Algorithm-Kernel Synergy [32.21687694159134]
  FluxBin is an algorithm that synerizes binary quantization with a highly optimized kernel.<n>FluxBin achieves up to $5.92times speedup and $10.19times energy savings.<n>This effectively enables the deployment of 70B-scale models on one single A100 bases with a $4times memory reduction.
  arXiv  Detail & Related papers  (2026-08-16T08:01:06Z)
• TELLER: Non-intrusive Cross-Layer Root-Cause Analysis for LLM Inference [47.62941623025392]
  We present TELLER, a non-intrusive and Log-aware LLM inference Root-cause analysis framework.<n>TELLER collects NVTX/CUPTI traces and service logs without modifying model binaries.<n>It reconstructs per-request call-chain trees and aligns log lines with corresponding execution steps.<n>On top of these representations, TELLER combines numeric candidate localization with a multimodal root-cause model.
  arXiv  Detail & Related papers  (2026-08-03T09:39:11Z)
• Constrained Decoding for Diffusion Language Models via Efficient Inference over Finite Automata [57.27430779838529]
  We present an exact and tractable algorithm for sampling from the constrained mean-field posterior under any constraint expressible as a finite automaton.<n>The approach guarantees constraint satisfaction by construction, supports both greedy and sampling-based decoding, and is compatible with parallel and block-wise decoding.<n> Empirical evaluations on Dream-7B and LLaDA-8 show substantial accuracy gains across various tasks.
  arXiv  Detail & Related papers  (2026-07-08T05:48:57Z)
• An Efficient vLLM-Based Inference Pipeline for Unified Audio Understanding and Generation [63.50393485685279]
  We present a vLLM-based inference pipeline for unified speech understanding and generation.<n>We overcome the shared intuition-Free Guidance (CFG) halves throughput.<n>By co-scheduling paired conditional and unconditional requests within a continuous batch, our CFG implementation sustains 80% of non-CFG throughput.
  arXiv  Detail & Related papers  (2026-07-02T12:55:11Z)
• Accelerating Discrete Diffusion Models with Parallel-In-Time Sampling [55.388363730120325]
  We parallelize the mainstream $$-leaping algorithm for absorbing discrete diffusion in a Continuous-Time Markov Chain (CTMC) framework.<n>By leveraging the continuous-time integral form of the $$-leaping algorithm and the Picard method, we achieve parallel-in-time sampling acceleration.<n>Our research expands the potential of discrete diffusion models for efficient parallel inference, with broader implications for applications such as molecular structure and language generation.
  arXiv  Detail & Related papers  (2026-07-01T10:59:33Z)
• Closed-Form Spectral Regularization for Multi-Task Model Merging [96.82449201305234]
  Model merging combines several independently fine-tuned experts into a single multi-task model without any training data.<n>State-of-the-art merging methods formulate merging as a layer-wise interference problem.<n>We propose SWUDI, a closed-form method that combines a soft exponential filter, which matches the gradient-flow trajectory of iterative descent.
  arXiv  Detail & Related papers  (2026-06-05T14:00:47Z)
• FlashSampling: Fast and Memory-Efficient Exact Sampling [62.5203057469482]
  FlashSampling is an exact sampling primitive that fuses sampling into the LM-head matmul and never materializes the logits tensor in.<n>Across H100, H200, B200, and B300 GPUs, FlashSampling speeds up kernel-level decode workloads.<n>In end-to-end vLLM experiments, it reduces time per output token by up to $19%$ on the models we test.
  arXiv  Detail & Related papers  (2026-03-16T19:37:08Z)
• Hardware-Agnostic Modeling of Quantum Side-Channel Leakage via Conditional Dynamics and Learning from Full Correlation Data [0.0]
  We study a sequential side-channel model in which an adversarial probe qubit interacts with a target qubit during a hidden gate sequence.<n>Repeating the same hidden sequence for $N$ shots yields an empirical full-correlation record.<n>Experiments over broad coupling and noise grids show that strict sequence recovery concentrates near the predicted coupling band.
  arXiv  Detail & Related papers  (2026-02-17T19:33:23Z)
• Parallel Sampling via Autospeculation [13.643401888306398]
  We present algorithms to accelerate sampling via counting in two settings: any-order autoregressive models and denoising diffusion models.<n>We show that, by issuing oracle calls in parallel, the expected sampling time can be reduced to $widetildeO(n1/2)$.<n>We introduce a novel technique to obtain our results: speculative rejection sampling.
  arXiv  Detail & Related papers  (2025-11-11T06:09:44Z)
• Learning to Parallel: Accelerating Diffusion Large Language Models via Learnable Parallel Decoding [21.609237262034636]
  Autoregressive decoding in large language models (LLMs) requires $mathcalO(n)$ sequential steps for $n$ tokens.<n>We propose Learning to Parallel Decode (Learn2PD), a framework that trains a lightweight and adaptive filter model to predict, for each token position, whether the current prediction matches the final output.<n>This learned filter approximates an oracle parallel decoding strategy that unmasks tokens only when correctly predicted.
  arXiv  Detail & Related papers  (2025-09-29T17:59:54Z)
• Boundary-aware Decoupled Flow Networks for Realistic Extreme Rescaling [49.215957313126324]
  Recently developed generative methods, including invertible rescaling network (IRN) based and generative adversarial network (GAN) based methods, have demonstrated exceptional performance in image rescaling. However, IRN-based methods tend to produce over-smoothed results, while GAN-based methods easily generate fake details. We propose Boundary-aware Decoupled Flow Networks (BDFlow) to generate realistic and visually pleasing results.
  arXiv  Detail & Related papers  (2024-05-05T14:05:33Z)
• Distributed Extra-gradient with Optimal Complexity and Communication Guarantees [60.571030754252824]
  We consider monotone variational inequality (VI) problems in multi-GPU settings where multiple processors/workers/clients have access to local dual vectors. Extra-gradient, which is a de facto algorithm for monotone VI problems, has not been designed to be communication-efficient. We propose a quantized generalized extra-gradient (Q-GenX), which is an unbiased and adaptive compression method tailored to solve VIs.
  arXiv  Detail & Related papers  (2023-08-17T21:15:04Z)
• Non-Adaptive Adaptive Sampling on Turnstile Streams [57.619901304728366]
  We give the first relative-error algorithms for column subset selection, subspace approximation, projective clustering, and volume on turnstile streams that use space sublinear in $n$. Our adaptive sampling procedure has a number of applications to various data summarization problems that either improve state-of-the-art or have only been previously studied in the more relaxed row-arrival model.
  arXiv  Detail & Related papers  (2020-04-23T05:00:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.