Render to Reason: Novel-View Semantic Prediction Improves Spatial Understanding in VLMs
cs.CV
Submitted: 2026-10-04
Updated: 2026-10-10
Project page: https://yuqunw.github.io/Render2Reason
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Qwen3-VL Technical Report
- DepthLM: Metric Depth From Vision Language Models
- VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- Omni-View: Unlocking How Generation Facilitates Understanding in Unified 3D Model based on Multiview images
- OmniSpatial: Towards Comprehensive Spatial Reasoning Benchmark for Vision Language Models
- LLaVA-OneVision: Easy Visual Task Transfer
- SQA3D: Situated Question Answering in 3D Scenes
- SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features
- InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
- Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence
- VP-Bench: A Comprehensive Benchmark for Visual Prompting in Multimodal Large Language Models
- Cambrian-S: Towards Spatial Supersensing in Video
- From Flatland to Space: Teaching Vision-Language Models to Perceive and Reason in 3D
- SSR: Pushing the Limit of Spatial Intelligence with Structured Scene Reasoning
- ReVSI: Rebuilding Visual Spatial Intelligence Evaluation for Accurate Assessment of VLM 3D Reasoning
- SpaceMind: Camera-Guided Modality Fusion for Spatial Reasoning in Vision-Language Models
- How to Teach Large Multimodal Models New Skills
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