Large language models (LLMs) excel across many tasks, yet inference is still dominated by strictly token-by-token autoregression. Existing acceleration methods largely patch this pipeline and miss core human-reading ingredients: content-adaptive foresight, chunk-structure-aware compute allocation, and train--test consistency for preview/skimming. We propose the \textbf{Fovea-Block-Skip Transformer} (FBS), which injects a causal, trainable loop into Transformers via Parafovea-Attention Window (PAW), Chunk-Head (CH), and Skip-Gate (SG). Across diverse benchmarks, FBS improves the quality-efficiency trade-off without increasing parameters, and ablations show the three modules are complementary.

翻译：大型语言模型（LLMs）在众多任务中表现出色，但其推理过程仍主要受限于严格的逐词自回归生成。现有的加速方法大多是对这一流程的修补，未能捕捉人类阅读的核心要素：内容自适应前瞻、块结构感知的计算分配，以及预览/略读的训练-测试一致性。我们提出\textbf{Fovea-Block-Skip Transformer}（FBS），该模型通过旁中央凹注意力窗口（PAW）、块头（CH）和跳过门（SG）向Transformer注入了一个因果、可训练的循环机制。在多样化基准测试中，FBS在不增加参数的情况下改善了质量-效率权衡，消融实验表明三个模块具有互补性。