mlsanityData to 14 Oct 2026

TPP-SD: Accelerating Transformer Point Process Sampling with Speculative Decoding

arXiv 2507.09252 ↗First posted 12 Jul 2025cs.LG stat.MLcited by 3

On the map: Efficient LLM inference and serving → Speculative decoding

Abstract (arXiv)

We propose TPP-SD, a novel approach that accelerates Transformer temporal point process (TPP) sampling by adapting speculative decoding (SD) techniques from language models. By identifying the structural similarities between thinning algorithms for TPPs and speculative decoding for language models, we develop an efficient sampling framework that leverages a smaller draft model to generate multiple candidate events, which are then verified by the larger target model in parallel. TPP-SD maintains the same output distribution as autoregressive sampling while achieving significant acceleration. Experiments on both synthetic and real datasets demonstrate that our approach produces samples from identical distributions as standard methods, but with 2-6× speedup. Our ablation studies analyze the impact of hyperparameters such as draft length and draft model size on sampling efficiency. TPP-SD bridges the gap between powerful Transformer TPP models and the practical need for rapid sequence sampling.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Placed in Speculative decoding, one of 12 directions in Efficient LLM inference and serving. Builds on 15 of its 40 references that are in the map. First posted on arXiv 12 Jul 2025.

Builds on

15 of 40 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty26 Jan
  2. 2024Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads19 Jan
  3. 2023Accelerating Large Language Model Decoding with Speculative Sampling2 Feb
  4. 2022Fast Inference from Transformers via Speculative Decoding30 Nov

Other references · 11

  1. 2022Efficiently Scaling Transformer Inferencecited by 539
  2. 2020Transformer Hawkes Processcited by 235
  3. 2019Intensity-Free Learning of Temporal Point Processescited by 140
  4. 2021Transformer Embeddings of Irregularly Spaced Events and Their Participantscited by 93
  5. 2022HYPRO: A Hybridly Normalized Probabilistic Model for Long-Horizon Prediction of Event Sequencescited by 57
  6. 2018Lecture Notes: Temporal Point Processes and the Conditional Intensity Functioncited by 40
  7. 2020Fast and Flexible Temporal Point Processes with Triangular Mapscited by 30
  8. 2024Continuous Speculative Decoding for Autoregressive Image Generationcited by 19
  9. 2024Decomposable Transformer Point Processescited by 16
  10. 2024Interpretable Transformer Hawkes Processes: Unveiling Complex Interactions in Social Networkscited by 11
  11. 2024TransFeat-TPP: An Interpretable Deep Covariate Temporal Point Processescited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2026Make Every Draft Count: Hidden State based Speculative Decodingcited by 4
  2. 2025Accelerating Time Series Foundation Models with Speculative Decodingcited by 1
  3. 2026One Model, Multiple Compute Budgets: Multi-Ratio ConceptMoE for Flexible Compute Across Turnscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024A Theoretical Perspective for Speculative Decoding Algorithm
  2. 2025Fast Inference via Hierarchical Speculative Decoding
  3. 2026D^2SD: Accelerating Speculative Decoding with Dual Diffusion Draft Models
  4. 2026LongSpark: Efficient speculative decoding with a fixed-cost parallel drafter
  5. 2024Speculative Diffusion Decoding: Accelerating Language Generation through Diffusion
  6. 2025DiffuSpec: Unlocking Diffusion Language Models for Speculative Decoding
  7. 2025SpecDiff-2: Scaling Diffusion Drafter Alignment For Faster Speculative Decoding
  8. 2025Speculative Sampling for Parametric Temporal Point Processes