Rotated, but How Far? Diagnosing and Improving Object-Rotation Reasoning in VLMs
cs.CV, cs.AI
Submitted: 2026-10-05
Updated: 2026-10-05
Sources
- Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
- Cosmos-Reason1: From Physical Common Sense To Embodied Reasoning
- Qwen3-VL Technical Report
- Qwen2.5-VL Technical Report
- Linear Mechanisms for Spatiotemporal Reasoning in Vision Language Models
- LRR-Bench: Left, Right or Rotate? Vision-Language models Still Struggle With Spatial Understanding Tasks
- LLaVA-OneVision: Easy Visual Task Transfer
- Diagnosing Bottlenecks in Data Visualization Understanding by Vision-Language Models
- Probing Visual Concepts in Lightweight Vision-Language Models for Automated Driving
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
- InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
- InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models
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