Moving Off-the-Grid: Scene-Grounded Video Representations
–Neural Information Processing Systems
Current vision models typically maintain a fixed correspondence between their representation structure and image space. Each layer comprises a set of tokens arranged "on-the-grid," which biases patches or tokens to encode information at
Neural Information Processing Systems
Oct-10-2025, 19:12:42 GMT
- Genre:
- Research Report > Experimental Study (0.93)
- Industry:
- Energy > Power Industry (0.41)
- Information Technology (0.92)
- Technology: