提出高效3D多模态模型推理框架，动态剪枝视觉令牌以提速保准。
Efficient3D: A Unified Framework for Adaptive and Debiased Token Reduction in 3D MLLMs
- 设计去偏重要性评估模块，提升视觉令牌重要性预测可靠性。
- 引入自适应重平衡策略，按场景复杂度动态调整剪枝强度。
- 在5个3D基准上验证，显著提速且关键指标提升2.57%。
近年来，多模态大语言模型（MLLMs）在3D领域拓展了推理能力，实现了细粒度空间理解。然而，3D MLLMs规模庞大且输入特征维度高，带来显著推理开销，限制了在资源受限平台的实际部署。为此，本文提出Efficient3D，一个统一的视觉令牌剪枝框架，可在保持竞争性准确率的同时加速3D MLLMs推理。该框架引入去偏视觉令牌重要性估计器（DVTIE），考虑浅层初始层对注意力聚合的影响，从而生成更可靠的视觉令牌重要性预测。同时，提出自适应令牌重平衡（ATR）策略，根据场景复杂度动态调整剪枝强度，确保语义完整性并维持各层注意力平衡。二者协同实现上下文感知的令牌缩减，以更低计算量保留关键语义。在五个代表性3D视觉与语言基准（ScanRefer、Multi3DRefer、Scan2Cap、ScanQA、SQA3D）上的全面实验表明，Efficient3D优于未剪枝基线，在Scan2Cap数据集上取得+2.57%的CIDEr提升。因此，Efficient3D为3D MLLMs的高效推理提供了可扩展、有效的解决方案。代码已开源：https://github.com/sol924/Efficient3D。
原文摘要 · Abstract (English)
Recent advances in Multimodal Large Language Models (MLLMs) have expanded reasoning capabilities into 3D domains, enabling fine-grained spatial understanding. However, the substantial size of 3D MLLMs and the high dimensionality of input features introduce considerable inference overhead, which limits practical deployment on resource constrained platforms. To overcome this limitation, this paper presents Efficient3D, a unified framework for visual token pruning that accelerates 3D MLLMs while maintaining competitive accuracy. The proposed framework introduces a Debiased Visual Token Importance Estimator (DVTIE) module, which considers the influence of shallow initial layers during attention aggregation, thereby producing more reliable importance predictions for visual tokens. In addition, an Adaptive Token Rebalancing (ATR) strategy is developed to dynamically adjust pruning strength based on scene complexity, preserving semantic completeness and maintaining balanced attention across layers. Together, they enable context-aware token reduction that maintains essential semantics with lower computation. Comprehensive experiments conducted on five representative 3D vision and language benchmarks, including ScanRefer, Multi3DRefer, Scan2Cap, ScanQA, and SQA3D, demonstrate that Efficient3D achieves superior performance compared with unpruned baselines, with a +2.57% CIDEr improvement on the Scan2Cap dataset. Therefore, Efficient3D provides a scalable and effective solution for efficient inference in 3D MLLMs. The code is released at: https://github.com/sol924/Efficient3D
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。