UniPrefill: Universal Long-Context Prefill Acceleration via Block-wise Dynamic Sparsification

arXiv:2605.06221 · cs.CL · Submitted 2026-05-07 · Read on arXiv

cs.CL

Submitted: 2026-05-07

Updated: 2026-09-25

Code: https://github.com/qhfan/UniPrefill

Terminology

• Sparse Attention — 15× in this paper · explained in 5 episodes
• Accuracy — 14× in this paper · explained in 1 episode
• Decode — 11× in this paper · explained in 2 episodes
• Hybrid Architectures — 10× in this paper · explained in 3 episodes
• Transformer — 8× in this paper · explained in 2 episodes
• Ablation — 6× in this paper · explained in 1 episode
• Sparsity — 6× in this paper · explained in 4 episodes
• Benchmark — 5× in this paper · explained in 2 episodes
• KV cache — 5× in this paper · explained in 3 episodes
• Ablation study — 4× in this paper · explained in 1 episode
• Large Language Models (LLMs) — 3× in this paper · explained in 6 episodes
• Attention Mechanism — 2× in this paper · explained in 2 episodes
• Clustering — 2× in this paper · explained in 2 episodes
• Pruning — 2× in this paper · explained in 4 episodes
• Ablation Studies — 1× in this paper · explained in 1 episode
• Aggregation — 1× in this paper · explained in 1 episode
• Alignment — 1× in this paper · explained in 1 episode
• Allocator — 1× in this paper · explained in 1 episode
• Consistency — 1× in this paper · explained in 3 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 2 episodes
• Hybrid Models — 1× in this paper · explained in 2 episodes

Sources

• Qwen Technical Report
• VSPrefill: Vertical-Slash Sparse Attention with Lightweight Indexing for Long-Context Prefilling
• The Llama 3 Herd of Models
• FlashPrefill: Instantaneous Pattern Discovery and Thresholding for Ultra-Fast Long-Context Prefilling
• LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference
• Mamba: Linear-Time Sequence Modeling with Selective State Spaces
• RULER: What's the Real Context Size of Your Long-Context Language Models?
• Mistral 7B
• Gemma 3 Technical Report
• MiniMax-01: Scaling Foundation Models with Lightning Attention
• SnapKV: LLM Knows What You are Looking for Before Generation
• MoBA: Mixture of Block Attention for Long-Context LLMs
• Retentive Network: A Successor to Transformer for Large Language Models
• LLaMA: Open and Efficient Foundation Language Models
• Llama 2: Open Foundation and Fine-Tuned Chat Models
• ProxyAttn: Guided Sparse Attention via Representative Heads
• MiMo-V2-Flash Technical Report
• Optimizing Mixture of Block Attention
• Qwen3 Technical Report
• Qwen2.5 Technical Report

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering