{"ID":2878344,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.17595","arxiv_id":"2508.17595","title":"TinyGiantVLM: A Lightweight Vision-Language Architecture for Spatial Reasoning under Resource Constraints","abstract":"Reasoning about fine-grained spatial relationships in warehouse-scale environments poses a significant challenge for existing vision-language models (VLMs), which often struggle to comprehend 3D layouts, object arrangements, and multimodal cues in real-world industrial settings. In this paper, we present TinyGiantVLM, a lightweight and modular two-stage framework designed for physical spatial reasoning, distinguishing itself from traditional geographic reasoning in complex logistics scenes. Our approach encodes both global and region-level features from RGB and depth modalities using pretrained visual backbones. To effectively handle the complexity of high-modality inputs and diverse question types, we incorporate a Mixture-of-Experts (MoE) fusion module, which dynamically combines spatial representations to support downstream reasoning tasks and improve convergence. Training is conducted in a two-phase strategy: the first phase focuses on generating free-form answers to enhance spatial reasoning ability, while the second phase uses normalized answers for evaluation. Evaluated on Track 3 of the AI City Challenge 2025, our 64M-parameter base model achieved 5th place on the leaderboard with a score of 66.8861, demonstrating strong performance in bridging visual perception and spatial understanding in industrial environments. We further present an 80M-parameter variant with expanded MoE capacity, which demonstrates improved performance on spatial reasoning tasks.","short_abstract":"Reasoning about fine-grained spatial relationships in warehouse-scale environments poses a significant challenge for existing vision-language models (VLMs), which often struggle to comprehend 3D layouts, object arrangements, and multimodal cues in real-world industrial settings. In this paper, we present TinyGiantVLM,...","url_abs":"https://arxiv.org/abs/2508.17595","url_pdf":"https://arxiv.org/pdf/2508.17595v1","authors":"[\"Vinh-Thuan Ly\",\"Hoang M. Truong\",\"Xuan-Huong Nguyen\"]","published":"2025-08-25T01:36:22Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Language Model\"]","has_code":false}