多模态大模型推理成本主要来自视觉token，提出逐步选择方法大幅降低计算量，实现高效生成。
arXiv:2606.16067v1 Announce Type: new Abstract: In multimodal large language models (MLLMs), inference cost is largely dominated by the visual token p…
深入剖析注意力机制中标准化的局限性，揭示模型选择性能力与token选择的几何分离理论边界。
arXiv:2508.17821v3 Announce Type: replace-cross Abstract: This paper investigates the limitations of the normalization in attention mechanisms. We beg…
RL优化token选择，提升代码水印隐蔽性与鲁棒性，ICML 2026前沿研究揭秘新范式
arXiv:2508.11925v3 Announce Type: replace-cross Abstract: Protecting intellectual property on LLM-generated code necessitates effective watermarking s…
提出训练轨迹感知的token选择方法，为大模型训练效率带来新突破，已被ICML 2026接收。
arXiv:2601.10348v2 Announce Type: replace Abstract: Efficient distillation is a key pathway for converting expensive reasoning capability into deploya…
阶段自适应token选择策略，显著提升全模态大语言模型推理效率，突破多任务性能瓶颈。
arXiv:2605.20035v1 Announce Type: new Abstract: Omni-modal large language models (om-LLMs) achieve unified audio-visual understanding by encoding vide…