# AsyncSpade: Efficient Test-Time Scaling with Asynchronous Sparse Decoding

arXiv: https://arxiv.org/abs/2510.07486 · HTML: https://mlsanity.com/paper/2510.07486/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 16 of its 42 references that are in the map.
- First posted on arXiv 8 Oct 2025.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)
- Efficient LLM inference and serving / [Speculative decoding](https://mlsanity.com/inference/speculative-decoding/)

## Builds on (most cited first, 16 of 16)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [Kimi K2: Open Agentic Intelligence](https://mlsanity.com/paper/2507.20534/) (2025)
- [LIMO: Less is More for Reasoning](https://mlsanity.com/paper/2502.03387/) (2025)
- [DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition](https://mlsanity.com/paper/2504.21801/) (2025)
- [A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?](https://mlsanity.com/paper/2503.24235/) (2025)
- [MoBA: Mixture of Block Attention for Long-Context LLMs](https://mlsanity.com/paper/2502.13189/) (2025)
- [Kimina-Prover Preview: Towards Large Formal Reasoning Models with Reinforcement Learning](https://mlsanity.com/paper/2504.11354/) (2025)
- [Transformers are Multi-State RNNs](https://mlsanity.com/paper/2401.06104/) (2024)
- [OpenCodeReasoning: Advancing Data Distillation for Competitive Coding](https://mlsanity.com/paper/2504.01943/) (2025)
- [MedReason: Eliciting Factual Medical Reasoning Steps in LLMs via Knowledge Graphs](https://mlsanity.com/paper/2504.00993/) (2025)
- [Winning Gold at IMO 2025 with a Model-Agnostic Verification-and-Refinement Pipeline](https://mlsanity.com/paper/2507.15855/) (2025)
- [Inference-Time Hyper-Scaling with KV Cache Compression](https://mlsanity.com/paper/2506.05345/) (2025)

## Built on by (most cited first, 2 of 2)

- [Predict, Reuse, and Repair: Accelerating Dynamic Sparse Attention for Long-Context LLM Decoding](https://mlsanity.com/paper/2606.30389/) (2026)
- [CoSA: Accelerating Long-Context Inference via Proxy-Kernel Co-Designed Sparse Attention](https://mlsanity.com/paper/2607.25291/) (2026)

## Abstract (arXiv)

Test-time scaling (TTS) boosts LLM reasoning via long chain-of-thought (CoT), but the linear KV-cache growth amplifies the memory-bound bottleneck of LLM decoding. Query-aware page-level sparse decoding can achieve state-of-the-art performance under constrained FLOPs budgets, but is limited by both sequential-dependent page filtering and coarse-grained token selection, hampering serving efficiency and model performance on TTS tasks under high concurrency and long CoT scenarios (consuming even higher runtime than the forward pipeline itself). In this paper, we first find that the current-step query state can be accurately approximated in a unified manner from a short window of recent queries, enabling training-free query-aware sparsity without waiting in the decoding loop. We propose AsyncSpade, an asynchronous framework for efficient TTS built on two core components: (1) a novel light-weight temporal-regressive module that predicts the next-token query state; (2) an asynchronous and disaggregated framework that decouples the KV cache filtering from the auto-regressive decoding loop, overlapping the token-level KV selection with the forward inference computation through asynchronism. To our knowledge, AsyncSpade is the first to eliminate the sequential dependence without sacrificing model performance. We validate the effectiveness of AsyncSpade on common LLM serving setups with an A100 node, where AsyncSpade fully overlaps KV-cache operations with the inference pipeline, achieving theoretical optimal time-per-output-token (TPOT). Specifically, AsyncSpade delivers over 20% reduction on TPOT compared to SoTA baseline (i.e. Quest) and at least 50% TPOT reduction compared to full attention on Qwen3-8B and Qwen3-32B models, while matching or surpassing their accuracy on various TTS benchmarks (AIME-24/25, GPQA-Diamond, MATH-500).