mlsanityData to 7 Oct 2026

PACER: Blockwise Pre-verification for Speculative Decoding with Adaptive Length

arXiv 2602.01274 ↗First posted 1 Feb 2026cs.CL cs.AIcited by 2

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

Abstract (arXiv)

Speculative decoding (SD) is a powerful technique for accelerating the inference process of large language models (LLMs) without sacrificing accuracy. Typically, SD employs a small draft model to generate a fixed number of draft tokens, which are then verified in parallel by the target model. However, our experiments reveal that the optimal draft length varies significantly across different decoding steps. This variation suggests that using a fixed draft length limits the potential for further improvements in decoding speed. To address this challenge, we propose Pacer, a novel approach that dynamically controls draft length using a lightweight, trainable pre-verification layer. This layer pre-verifies draft tokens blockwise before they are sent to the target model, allowing the draft model to stop token generation if the blockwise pre-verification fails. We implement Pacer on multiple SD model pairs and evaluate its performance across various benchmarks. Our results demonstrate that Pacer achieves up to 2.66x Speedup over autoregressive decoding and consistently outperforms standard speculative decoding. Furthermore, when integrated with Ouroboros, Pacer attains up to 3.09x Speedup.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Speculative decoding, one of 12 directions in Efficient LLM inference and serving. Builds on 58 of its 67 references that are in the map. First posted on arXiv 1 Feb 2026.

Builds on

58 of 67 references.

Closest ancestors · 26

Same direction, newest first

  1. 2025Token-Driven GammaTune: Adaptive Calibration for Enhanced Speculative Decoding28 Mar
  2. 2025C2T: A Classifier-Based Tree Construction Method in Speculative Decoding19 Feb
  3. 2024Draft Model Knows When to Stop: Self-Verification Speculative Decoding for Long-Form Generation27 Nov
  4. 2024AdaEDL: Early Draft Stopping for Speculative Decoding of Large Language Models via an Entropy-based Lower Bound on Token Acceptance Probability24 Oct
  5. 2024ParallelSpec: Parallel Drafter for Efficient Speculative Decoding8 Oct
  6. 2024Dynamic Depth Decoding: Faster Speculative Decoding for LLMs30 Aug
  7. 2024PEARL: Parallel Speculative Decoding with Adaptive Draft Length13 Aug
  8. 2024OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure25 Jun
  9. 2024EAGLE-2: Faster Inference of Language Models with Dynamic Draft Trees24 Jun
  10. 2024SpecDec++: Boosting Speculative Decoding via Adaptive Candidate Lengths30 May
  11. 2024Dynamic Speculation Lookahead Accelerates Speculative Decoding of Large Language Models7 May
  12. 2024Ouroboros: Generating Longer Drafts Phrase by Phrase for Faster Speculative Decoding21 Feb
14 more
  1. 2024Generation Meets Verification: Accelerating Large Language Model Inference with Smart Parallel Auto-Correct Decoding19 Feb
  2. 2024Break the Sequential Dependency of LLM Inference Using Lookahead Decoding3 Feb
  3. 2024EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty26 Jan
  4. 2024Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads19 Jan
  5. 2024Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding15 Jan
  6. 2023REST: Retrieval-Based Speculative Decoding14 Nov
  7. 2023SpecTr: Fast Speculative Decoding via Optimal Transport23 Oct
  8. 2023DistillSpec: Improving Speculative Decoding via Knowledge Distillation12 Oct
  9. 2023Accelerating LLM Inference with Staged Speculative Decoding8 Aug
  10. 2023Accelerating Transformer Inference for Translation via Parallel Decoding17 May
  11. 2023SpecInfer: Accelerating Generative Large Language Model Serving with Tree-based Speculative Inference and Verification16 May
  12. 2023Accelerating Large Language Model Decoding with Speculative Sampling2 Feb
  13. 2022Fast Inference from Transformers via Speculative Decoding30 Nov
  14. 2022Speculative Decoding: Exploiting Speculative Execution for Accelerating Seq2seq Generation30 Mar

Other references · 32

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  3. 2025Qwen3 Technical Reportcited by 8,195
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  6. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  7. 2024Qwen2.5 Technical Reportcited by 4,713
  8. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,611
  9. 2021Program Synthesis with Large Language Modelscited by 3,509
  10. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  11. 2023FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioningcited by 2,551
  12. 2023Efficient Streaming Language Models with Attention Sinkscited by 2,115
20 more
  1. 2017Get To The Point: Summarization with Pointer-Generator Networkscited by 1,281
  2. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,234
  3. 2024DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligencecited by 1,154
  4. 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,151
  5. 2024SnapKV: LLM Knows What You are Looking for Before Generationcited by 848
  6. 2023H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Modelscited by 805
  7. 2019Fast Transformer Decoding: One Write-Head is All You Needcited by 700
  8. 2024Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inferencecited by 487
  9. 2023Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Timecited by 428
  10. 2025Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attentioncited by 391
  11. 2024Better & Faster Large Language Models via Multi-token Predictioncited by 344
  12. 2018Blockwise Parallel Decoding for Deep Autoregressive Modelscited by 318
  13. 2024LayerSkip: Enabling Early Exit Inference and Self-Speculative Decodingcited by 302
  14. 2024MagicPIG: LSH Sampling for Efficient LLM Generationcited by 112
  15. 2020Data Movement Is All You Need: A Case Study on Optimizing Transformerscited by 108
  16. 2025Towards Robust Mathematical Reasoningcited by 67
  17. 2024Hydra: Bidirectional State Space Models Through Generalized Matrix Mixerscited by 43
  18. 2023Predictive Pipelined Decoding: A Compute-Latency Trade-off for Exact LLM Decodingcited by 40
  19. 2024Detecting Hallucinations in Large Language Model Generation: A Token Probability Approachcited by 40
  20. 2024Cost-Effective Hallucination Detection for LLMscited by 16

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Fully Rotation-Equivariant Spectral-Spatial Learning for Multispectral Object Detectioncited by 0
  2. 2026DLoop: Looped Speculative Decodingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Block Verification Accelerates Speculative Decoding
  2. 2026LongSpark: Efficient speculative decoding with a fixed-cost parallel drafter
  3. 2025Speeding up Speculative Decoding via Sequential Approximate Verification
  4. 2026SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding
  5. 2026SpecTr-GBV: Multi-Draft Block Verification Accelerating Speculative Decoding
  6. 2026DBLAST: Dependent Block Drafting for Stochastic Speculative Decoding
  7. 2025Speculative Decoding: Performance or Illusion?
  8. 2024Decoding Speculative Decoding