SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features | VitaVision
PaperarXiv preprint · 2025

SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features

Michael TschannenAlexey A. GritsenkoXiao WangMuhammad Ferjad NaeemIbrahim AlabdulmohsinNikhil ParthasarathyTalfan EvansLucas BeyerYe XiaBasil MustafaOlivier J. HénaffJeremiah HarmsenAndreas SteinerXiaohua Zhai
arXiv 2502.14786DOI
0pages built on it
0more pages cite it
1narrative
0registry papers cite it

In the Atlas

No Atlas page is built on this paper yet.

Appears in 1 narrative

NarrativeFoundation Models for Vision

Lineage

Papers in the Atlas registry only · one dot per paper
Built upon by · 0
’20’21’22’23’24’25An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (2020)Learning Transferable Visual Models From Natural Language Supervision (2021)
Built upon by · 0