Token Sparse Attention: Efficient Long-Context Inference with Interleaved Token Selection

• URL: http://arxiv.org/abs/2602.03216v1
• Date: Tue, 03 Feb 2026 07:31:14 GMT
• Title: Token Sparse Attention: Efficient Long-Context Inference with Interleaved Token Selection
• Authors: Dongwon Jo, Beomseok Kang, Jiwon Song, Jae-Joon Kim,
• Abstract summary: Token Sparse Attention is a dynamic token-level sparsification mechanism that compresses per-head $Q$, $K$, $V$ to a reduced token set during attention.<n>We show that Token Sparse Attention consistently improves accuracy-latency trade-off, achieving up to $times$3.23 attention speedup at 128K context with less than 1% accuracy degradation.
• Score: 13.937483869660648
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The quadratic complexity of attention remains the central bottleneck in long-context inference for large language models. Prior acceleration methods either sparsify the attention map with structured patterns or permanently evict tokens at specific layers, which can retain irrelevant tokens or rely on irreversible early decisions despite the layer-/head-wise dynamics of token importance. In this paper, we propose Token Sparse Attention, a lightweight and dynamic token-level sparsification mechanism that compresses per-head $Q$, $K$, $V$ to a reduced token set during attention and then decompresses the output back to the original sequence, enabling token information to be reconsidered in subsequent layers. Furthermore, Token Sparse Attention exposes a new design point at the intersection of token selection and sparse attention. Our approach is fully compatible with dense attention implementations, including Flash Attention, and can be seamlessly composed with existing sparse attention kernels. Experimental results show that Token Sparse Attention consistently improves accuracy-latency trade-off, achieving up to $\times$3.23 attention speedup at 128K context with less than 1% accuracy degradation. These results demonstrate that dynamic and interleaved token-level sparsification is a complementary and effective strategy for scalable long-context inference.

Related papers

• SPOT: Sparsification with Attention Dynamics via Token Relevance in Vision Transformers [9.24617274894642]
  We present SParsification with attentiOn dynamics via Token relevance (SPOT) as a framework for early detection of redundant tokens within Vision Transformers (ViTs)<n>SPOT informs token sparsification and facilitates the elimination of such tokens, improving computational efficiency without sacrificing performance.<n> Empirical evaluations demonstrate significant efficiency gains of up to 40% compared to standard ViTs.
  arXiv  Detail & Related papers  (2025-11-13T16:56:24Z)
• DELTA: Dynamic Layer-Aware Token Attention for Efficient Long-Context Reasoning [6.468843780300177]
  We present textbfDELTA, a training-free sparse attention mechanism that achieves computational efficiency without sacrificing model accuracy.<n>Our results show that selective reuse of intermediate attention maps offers a robust path toward efficient long-context reasoning.
  arXiv  Detail & Related papers  (2025-10-10T21:37:49Z)
• Training-Free Token Pruning via Zeroth-Order Gradient Estimation in Vision-Language Models [16.540220733551823]
  Large Vision-Language Models (VLMs) enable strong multimodal reasoning but incur heavy inference costs from redundant visual tokens.<n> Attention-based methods rely on raw attention scores, which are often unstable across layers and heads.<n>We propose ours, a training-free framework built on a simple intuition.
  arXiv  Detail & Related papers  (2025-09-29T14:20:05Z)
• Multipole Attention for Efficient Long Context Reasoning [64.94673641704289]
  Large Reasoning Models (LRMs) have shown promising accuracy improvements on complex problem-solving tasks.<n>LRMs need to generate long chain-of-thought reasoning in order to think before answering.<n>We introduce Multipole Attention, which accelerates autoregressive reasoning by only computing exact attention for the most important tokens.
  arXiv  Detail & Related papers  (2025-06-16T03:00:40Z)
• ToDRE: Visual Token Pruning via Diversity and Task Awareness for Efficient Large Vision-Language Models [59.47738955960352]
  ToDRE is a two-stage and training-free token compression framework.<n>It achieves superior performance by pruning tokens based on token Diversity and token-task RElevance.
  arXiv  Detail & Related papers  (2025-05-24T15:47:49Z)
• TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model [56.43860351559185]
  We introduce textbfTopV, a compatible textbfTOken textbfPruning with inference Time Optimization for fast and low-memory textbfVLM.<n>Our framework incorporates a visual-aware cost function to measure the importance of each source visual token, enabling effective pruning of low-importance tokens.
  arXiv  Detail & Related papers  (2025-03-24T01:47:26Z)
• VQToken: Neural Discrete Token Representation Learning for Extreme Token Reduction in Video Large Language Models [35.38573641029626]
  We introduce the novel task of Extreme Short Token Reduction, which aims to represent entire videos using a minimal set of discrete tokens.<n>On the Extreme Short Token Reduction task, our VQToken compresses sequences to just 0.07 percent of their original length while incurring only a 0.66 percent drop in accuracy on the NextQA-MC benchmark.
  arXiv  Detail & Related papers  (2025-03-21T09:46:31Z)
• "Principal Components" Enable A New Language of Images [79.45806370905775]
  We introduce a novel visual tokenization framework that embeds a provable PCA-like structure into the latent token space.<n>Our approach achieves state-of-the-art reconstruction performance and enables better interpretability to align with the human vision system.
  arXiv  Detail & Related papers  (2025-03-11T17:59:41Z)
• Tactic: Adaptive Sparse Attention with Clustering and Distribution Fitting for Long-Context LLMs [10.52833484759311]
  We propose Tactic, a sparsity-adaptive and calibration-free sparse attention mechanism.<n>It dynamically selects tokens based on their cumulative attention scores rather than a fixed token budget.<n>We show that Tactic outperforms existing sparse attention algorithms, achieving superior accuracy and up to 7.29x decode attention speedup.
  arXiv  Detail & Related papers  (2025-02-17T08:39:43Z)
• AttentionPredictor: Temporal Patterns Matter for KV Cache Compression [64.75459635661562]
  We propose AttentionPredictor, which is the first learning-based method to directly predict attention patterns for KV cache compression and critical token identification.<n> AttentionPredictor accurately predicts the attention score and shares the unified prediction model, which consumes negligible memory.<n>By retaining most of the attention information, AttentionPredictor achieves 13$times$ KV cache compression and 5.6$times$ speedup in a cache offloading scenario.
  arXiv  Detail & Related papers  (2025-02-06T13:41:46Z)
• ToSA: Token Selective Attention for Efficient Vision Transformers [50.13756218204456]
  ToSA is a token selective attention approach that can identify tokens that need to be attended as well as those that can skip a transformer layer. We show that ToSA can significantly reduce computation costs while maintaining accuracy on the ImageNet classification benchmark.
  arXiv  Detail & Related papers  (2024-06-13T05:17:21Z)

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.