Metric-Bench: Exploring In-context Spatial Metric Reasoning in VLMs for Indoor Scenes

arXiv:2609.25841 · cs.CV, cs.MM · Submitted 2026-09-22 · Read on arXiv

cs.CV, cs.MM

Submitted: 2026-09-22

Updated: 2026-09-22

Terminology

• Benchmark — 32× in this paper · explained in 2 episodes
• Fine-Tuning — 19× in this paper · explained in 5 episodes
• Accuracy — 15× in this paper · explained in 1 episode
• Reinforcement Fine-Tuning — 8× in this paper
• Consistency — 6× in this paper · explained in 3 episodes
• Ablation — 4× in this paper · explained in 1 episode
• Binned Proximity — 4× in this paper
• Calibration — 4× in this paper · explained in 7 episodes
• GRPO — 4× in this paper · explained in 6 episodes
• Grounding — 4× in this paper · explained in 3 episodes
• Reinforcement Learning — 4× in this paper · explained in 4 episodes
• Exponential Precision — 3× in this paper
• Multimodal Reasoning — 3× in this paper · explained in 5 episodes
• Adaptation — 2× in this paper · explained in 1 episode
• CVF International Conference on Computer Vision — 2× in this paper
• Catastrophic Forgetting — 2× in this paper · explained in 5 episodes
• Chain-of-Thought (CoT) — 2× in this paper · explained in 4 episodes
• Core — 2× in this paper · explained in 2 episodes
• Foundation Model — 2× in this paper · explained in 3 episodes
• Generalization — 2× in this paper · explained in 13 episodes
• Reward Function — 2× in this paper · explained in 2 episodes
• Verifiable Rewards — 2× in this paper · explained in 2 episodes
• Verification — 2× in this paper · explained in 2 episodes
• Visual Grounding — 2× in this paper · explained in 2 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 3 episodes
• Chain-of-Thought (CoT) Prompting — 1× in this paper · explained in 3 episodes
• Chain-of-Thought Prompting — 1× in this paper · explained in 2 episodes
• Embodied AI — 1× in this paper · explained in 2 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Group Relative Policy Optimization — 1× in this paper
• Group Relative Policy Optimization (GRPO) — 1× in this paper · explained in 5 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• Mean Relative Accuracy — 1× in this paper
• Root Mean Squared Error — 1× in this paper
• Structured Reasoning — 1× in this paper · explained in 3 episodes
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes

Sources

• Qwen2.5-VL Technical Report
• DepthLM: Metric Depth From Vision Language Models
• Holistic Evaluation of Multimodal LLMs on Spatial Intelligence
• RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation
• Reasoning in Space via Grounding in the World
• 3D Aware Region Prompted Vision Language Model
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• Are Multimodal Large Language Models Ready for Omnidirectional Spatial Reasoning?
• BLINK: Multimodal Large Language Models Can See but Not Perceive
• Gemini Robotics: Bringing AI into the Physical World
• 3D-R1: Enhancing Reasoning in 3D VLMs for Unified Scene Understanding
• Enhancing Spatial Reasoning in Vision-Language Models via Chain-of-Thought Prompting and Reinforcement Learning
• OmniSpatial: Towards Comprehensive Spatial Reasoning Benchmark for Vision Language Models
• SPAZER: Spatial-Semantic Progressive Reasoning Agent for Zero-shot 3D Visual Grounding
• LLaVA-OneVision: Easy Visual Task Transfer
• PixelVLA: Advancing Pixel-level Understanding in Vision-Language-Action Model
• Reasoning Paths with Reference Objects Elicit Quantitative Spatial Reasoning in Large Vision-Language Models
• Improved Visual-Spatial Reasoning via R1-Zero-Like Training
• SSR: Enhancing Depth Perception in Vision-Language Models via Rationale-Guided Spatial Reasoning
• SpaceR: Reinforcing MLLMs in Video Spatial Reasoning

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models