PointCoT: A Multi-modal Benchmark for Explicit 3D Geometric Reasoning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918359967203328 |
|---|---|
| author | Zhang, Dongxu Sun, Yiding Li, Pengcheng Liu, Yumou Lin, Hongqiang Xu, Haoran Mu, Xiaoxuan Lin, Liang Yan, Wenbiao Yang, Ning Fang, Chaowei Zhao, Juanjuan Zhu, Jihua He, Conghui Tan, Cheng |
| author_facet | Zhang, Dongxu Sun, Yiding Li, Pengcheng Liu, Yumou Lin, Hongqiang Xu, Haoran Mu, Xiaoxuan Lin, Liang Yan, Wenbiao Yang, Ning Fang, Chaowei Zhao, Juanjuan Zhu, Jihua He, Conghui Tan, Cheng |
| contents | While Multimodal Large Language Models (MLLMs) demonstrate proficiency in 2D scenes, extending their perceptual intelligence to 3D point cloud understanding remains a significant challenge. Current approaches focus primarily on aligning 3D features with pre-trained models. However, they typically treat geometric reasoning as an implicit mapping process. These methods bypass intermediate logical steps and consequently suffer from geometric hallucinations. They confidently generate plausible responses that fail to ground in precise structural details. To bridge this gap, we present PointCoT, a novel framework that empowers MLLMs with explicit Chain-of-Thought (CoT) reasoning for 3D data. We advocate for a \textit{Look, Think, then Answer} paradigm. In this approach, the model is supervised to generate geometry-grounded rationales before predicting final answers. To facilitate this, we construct Point-Reason-Instruct, a large-scale benchmark comprising $\sim$86k instruction-tuning samples with hierarchical CoT annotations. By leveraging a dual-stream multi-modal architecture, our method synergizes semantic appearance with geometric truth. Extensive experiments demonstrate that PointCoT achieves state-of-the-art performance on complex reasoning tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_23945 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PointCoT: A Multi-modal Benchmark for Explicit 3D Geometric Reasoning Zhang, Dongxu Sun, Yiding Li, Pengcheng Liu, Yumou Lin, Hongqiang Xu, Haoran Mu, Xiaoxuan Lin, Liang Yan, Wenbiao Yang, Ning Fang, Chaowei Zhao, Juanjuan Zhu, Jihua He, Conghui Tan, Cheng Computer Vision and Pattern Recognition Artificial Intelligence Multimedia While Multimodal Large Language Models (MLLMs) demonstrate proficiency in 2D scenes, extending their perceptual intelligence to 3D point cloud understanding remains a significant challenge. Current approaches focus primarily on aligning 3D features with pre-trained models. However, they typically treat geometric reasoning as an implicit mapping process. These methods bypass intermediate logical steps and consequently suffer from geometric hallucinations. They confidently generate plausible responses that fail to ground in precise structural details. To bridge this gap, we present PointCoT, a novel framework that empowers MLLMs with explicit Chain-of-Thought (CoT) reasoning for 3D data. We advocate for a \textit{Look, Think, then Answer} paradigm. In this approach, the model is supervised to generate geometry-grounded rationales before predicting final answers. To facilitate this, we construct Point-Reason-Instruct, a large-scale benchmark comprising $\sim$86k instruction-tuning samples with hierarchical CoT annotations. By leveraging a dual-stream multi-modal architecture, our method synergizes semantic appearance with geometric truth. Extensive experiments demonstrate that PointCoT achieves state-of-the-art performance on complex reasoning tasks. |
| title | PointCoT: A Multi-modal Benchmark for Explicit 3D Geometric Reasoning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Multimedia |
| url | https://arxiv.org/abs/2602.23945 |