Research map / Architectures and efficiency / RNNs and normalization
AdaHC: Accelerating Multi-Token Prediction with Adaptive Head Chunking with Pipeline Parallelism
Yan Wang, Chang Si, Kaiming Yang, Zhipeng Zhang, Weijian Liu, Man Yuan, Mingzhen Li, Yong Li, Weile Jia
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add head chunking pipeline scheduling instead of naive tail placement | AdaHC · Qwen3-8B (no MTP) | throughput: +45.24% | Improved |
| use AdaHC instead of Naive with 1 MTP block | AdaHC · Qwen3-30B-A3B | throughput: +76.05% | Improved |
| use head chunking instead of layer redistribution | AdaHC · Qwen3-235B-A22B (1 MTP) | throughput: +22.93% | Improved |
| increase number of MTP blocks from 0 to 2 (ablation) | AdaHC · Qwen3-8B 29-layer | throughput gain over Naive: 21.56% -> 60.08% | Improved |
| use AdaHC instead of layer redistribution without MTP | AdaHC · Qwen3-8B 29-layer, no MTP | throughput | No clear effect |
| vocabulary-dimension partitioning instead of sequence-dimension chunking (ablation) | Vocabulary Parallel · Qwen3-8B no MTP | throughput: AdaHC +45.30% over VP | Hurt |
| move transformer layers to earlier stages (Layer Redistribute) | Naive · Qwen3-8B (no MTP) | throughput | Improved |
Closest papers by meaning
- TokenRouter: Efficient Serving System for Token-Level LLM Routing (NeurIPS 2026 · similarity 0.86)
- PipeOffload: Improving Scalability of Pipeline Parallelism with Memory Optimization (ICML 2025 · similarity 0.86)
- Synergistic Tensor and Pipeline Parallelism (NeurIPS 2025 · similarity 0.86)
- P-EAGLE: Parallel-Drafting EAGLE with Scalable Training (NeurIPS 2026 · similarity 0.85)
- AdaDecode: Accelerating LLM Decoding with Adaptive Layer Parallelism (ICML 2025 · similarity 0.85)
- Ada-K Routing: Boosting the Efficiency of MoE-based LLMs (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
