Soft Spatial Reasoning
cs.CV, cs.AI
Submitted: 2026-09-30
Updated: 2026-09-30
Code: https://github.com/rafiibnsultan/Soft_Spatial_Reasoning
Terminology
Sources
- Qwen Technical Report
- Qwen3-VL Technical Report
- SpatialThinker: Reinforcing Scene Graph-Grounded Spatial Reasoning via Dense Rewards
- DepthLM: Metric Depth From Vision Language Models
- Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas
- SIFThinker: Spatially-Aware Image Focus for Visual Reasoning
- Temperature as a Meta-Policy: Adaptive Temperature in LLM Reinforcement Learning
- Spatial Reasoning with Vision-Language Models in Ego-Centric Multi-View Scenes
- Training Large Language Models to Reason in a Continuous Latent Space
- CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughts
- G$^2$VLM: Geometry Grounded Vision Language Model with Unified 3D Reconstruction and Spatial Reasoning
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- GPT-4o System Card
- Vision-aligned Latent Reasoning for Multi-modal Large Language Model
- OmniSpatial: Towards Comprehensive Spatial Reasoning Benchmark for Vision Language Models
- Faithful GRPO: Improving Visual Spatial Reasoning in Multimodal Language Models via Constrained Policy Optimization
- SpatialLadder: Progressive Training for Spatial Reasoning in Vision-Language Models
- A Survey on Multimodal Benchmarks: In the Era of Large AI Models
- SSR: Enhancing Depth Perception in Vision-Language Models via Rationale-Guided Spatial Reasoning
- Thinking with Blueprints: Assisting Vision-Language Models in Spatial Reasoning via Structured Object Representation
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