Tuesday, October 27, 2026i10X News · Section desk8 stories
Studies
Multimodal
Vision, audio, video and multimodal model research
FilterGoogle
Filtered: Google in Multimodal · 2 storiesReset filter ×
-
Teaching LVLMs Spatial Skills With 15K Toy Blocks Beats Heavier Real-Scene Pipelines
SpatialBlock-15k: Qwen-based 4B model reaches 51.3% (+25.1 over backbone). 7B gains 17.6% over backbone; 3B beats SpatialLadder by 2.7–3.2% using only synthetic blocks.
By
Editor's picks
- Teaching LVLMs Spatial Skills With 15K Toy Blocks Beats Heavier Real-Scene Pipelines
- TurboVLA Runs Robot Policies at 32 Hz on an RTX 4090 Using Under 1 GB VRAM
- YuE2’s Symbolic Planning Wins 49.3% of Expert Preferences for Music Quality
- VLMs Still Can’t Tell Which Attribute Goes With Which Object — Until Text-Only Finetuning
- Qwen-Image-2.0 Unifies High-Fidelity Generation and Precise Editing in One Model
More stories
· RSS