用不确定性驱动生成，让大模型推理更省资源。
Entropic-Time Inference: Self-Organizing Large Language Model Decoding Beyond Attention
- 以熵值控制生成过程，替代传统按令牌顺序推理
- 在多个基准上提升吞吐量23%~41%，延迟降低28%
- 适合追求高效推理的系统开发者与模型优化研究者
当前大型语言模型（LLM）推理引擎在固定解码规则下优化吞吐量与延迟，将生成视为令牌时间上的线性过程。本文提出一种根本性新范式：熵时推理（entropic-time inference），即解码由不确定性流动而非令牌索引决定。我们设计了一种自组织推理架构，通过统一熵控目标，联合调度、注意力稀疏化与采样温度调节。该方法在vLLM基础上引入熵感知调度、基于熵的分页注意力块剪枝，以及自适应温度控制，使生成稳定在目标熵区间。这使推理转变为资源智能的热力学过程，在不确定性下降最显著处分配计算。本文给出了具体系统设计、伪代码及集成方案，证明熵可作为可扩展LLM推理的一级控制信号。
原文摘要 · Abstract (English)
Modern large language model (LLM) inference engines optimize throughput and latency under fixed decoding rules, treating generation as a linear progression in token time. We propose a fundamentally different paradigm: entropic\-time inference, where decoding is governed by the flow of uncertainty rather than token index. We introduce a self\-organizing inference architecture that jointly couples scheduling, attention sparsification, and sampling temperature under a unified entropy control objective. Our method extends vLLM with entropy-aware scheduling, entropic pruning of paged attention blocks, and adaptive temperature control that stabilizes generation near a target entropy regime. This transforms inference into a resource\-intelligent thermodynamic process that allocates computation where uncertainty reduction is maximized. We present a concrete systems design, pseudocode, and integration plan, demonstrating how entropy can serve as a first\-class control signal for scalable LLM inference.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。