HoPE: Hybrid of Position Embedding for Length Generalization in Vision-Language Models
Paper
•
2505.20444
•
Published
•
5
Haoran Li
brian13
AI & ML interests
Machine Learning, NLP, Multimodal Learning, Foundation Models
Recent Activity
upvoted
a
paper
about 23 hours ago
CoPE: Clipped RoPE as A Scalable Free Lunch for Long Context LLMs
authored
a paper
8 months ago
HoPE: Hybrid of Position Embedding for Length Generalization in
Vision-Language Models
upvoted
a
paper
8 months ago
HoPE: Hybrid of Position Embedding for Length Generalization in
Vision-Language Models
Organizations
None yet
authored
a
paper
8 months ago