Adaptive semantic alignment image captioning method
The method for image captioning based on frozen large language models breaks through the limitations of traditional models in utilizing external knowledge, but it tends to overly rely on textual priors, leading to insufficient use of visual features and hallucination issues in descriptions. To addre...
Furkejuvvon:
| Váldodahkkit: | , , , |
|---|---|
| Materiálatiipa: | Artigo |
| Giella: | Chinês |
| Almmustuhtton: |
National Computer System Engineering Research Institute of China
2026-05-01
|
| Ráidu: | Dianzi Jishu Yingyong |
| Fáttát: | |
| Liŋkkat: | http://www.chinaaet.com/article/3000177683 |
| Fáddágilkorat: |
Eai fáddágilkorat, Lasit vuosttaš fáddágilkora!
|