arXiv cs.CL· Duc Cao Dinh, Khai Le-Duc, Florent Draye, Chris Ngo, Terry Jingchen Zhang, Bernhard Sch\"olkopf, Zhijing Jin·· 12 小时前AI 评分20
PruneGround：面向 3D 视觉定位的即插即用空间剪枝框架
PruneGround: Plug-and-play Spatial Pruning for 3D Visual Grounding
AI 导读
PruneGround 是一个即插即用的 3D 视觉定位框架，先用冻结的 VLM 做语言引导空间剪枝（LGSP）缩小搜索区域，再通过多视角描述重构（MCDR）补全空间线索，并用 LLM-Grounder 将检测预训练空间 LLM 转为语言条件定位模型。在三个主流点云基准上，该方法在全部 ScanRefer 设置和 10 个 Nr3D/Sr3D 设置中的 9 个取得 SOTA，代码与模型已公开。
正文 · 原文
Abstract:3D Visual Grounding (3DVG) aims to localize target objects in 3D scenes given natural language descriptions. Existing approaches typically perform reasoning over the entire scene, leading to ambiguous predictions and high computational cost, especially in cluttered environments. We observe that many referential expressions rely on local spatial context and often correspond to restricted spatial regions rather than the full scene. Motivated by this insight, we propose PruneGround, an effective plug-and-play framework for 3DVG built upon three key components. First, we introduce Language-Guided Spatial Pruning (LGSP), which leverages a frozen Vision Language Model (VLM) to identify language-relevant regions, thereby reducing spatial computation and grounding candidates in the narrower search space. Second, we propose MultiView-Conditioned Description Reformulation (MCDR), which decomposes complex expressions into simplified target--anchor relations and augments missing spatial cues through multi-view reasoning. Finally, we propose LLM-Grounder, which repurposes a detection-pretrained spatial LLM into a language-conditioned grounding model by aligning point cloud and linguistic representations within the pruned region. Extensive experiments on the three most popular point cloud benchmarks demonstrate that our method achieves state-of-the-art results on all three ScanRefer settings and on 9 out of 10 Nr3D/Sr3D settings. Code and models are publicly available: this https URL
| Comments: | EMNLP 2026 (Oral) |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2606.31148 [cs.CV] |
| (or arXiv:2606.31148v3 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2606.31148 arXiv-issued DOI via DataCite |
Submission history
From: Khai Le-Duc [view email]
[v1]
Tue, 30 Jun 2026 05:21:11 UTC (16,845 KB)
[v2]
Fri, 4 Sep 2026 23:16:51 UTC (16,824 KB)
[v3]
Fri, 16 Oct 2026 06:53:09 UTC (16,824 KB)
来源：arXiv cs.CL · arxiv.org