Advancing Cross-domain Discriminability in Continual Learning of Vision-Language Models
–Neural Information Processing Systems
During the CL of VLMs, we need not only to prevent the catastrophic forgetting on incrementally learned knowledge but also to preserve the zero-shot ability of VLMs.
Neural Information Processing Systems
Feb-14-2026, 13:07:16 GMT
- Country:
- Asia
- China > Guangdong Province
- Guangzhou (0.04)
- Japan > Honshū
- Kantō > Tokyo Metropolis Prefecture > Tokyo (0.04)
- Middle East > Jordan (0.04)
- China > Guangdong Province
- North America > United States
- California
- Alameda County > Berkeley (0.04)
- San Francisco County > San Francisco (0.14)
- California
- Asia
- Genre:
- Research Report > Experimental Study (0.93)
- Industry:
- Information Technology (0.46)
- Technology:
- Information Technology
- Artificial Intelligence
- Machine Learning
- Neural Networks (1.00)
- Performance Analysis > Accuracy (0.47)
- Statistical Learning (0.67)
- Natural Language (1.00)
- Vision (1.00)
- Machine Learning
- Sensing and Signal Processing > Image Processing (1.00)
- Artificial Intelligence
- Information Technology