A lightweight explicit alignment recipe that adapts off-the-shelf VLMs into robust omni-modal embedding models. https://arxiv.org/abs/2601.03666
Haonan Chen
Haon-Chen
AI & ML interests
None yet
Recent Activity
authored a paper 1 day ago
Douyin Multimodal Embedding Model Technical Report upvoted a paper 5 days ago
UEmbed: Unified Sparse and Dense Multimodal Embeddings new activity 4 months ago
Haon-Chen/e5-omni-3B:Avoid trust_remote_code with transformers v5.6.0Organizations
models 5
Haon-Chen/e5-omni-3B
Visual Document Retrieval • 5B • Updated • 518 • 6
Haon-Chen/e5-omni-7B
Visual Document Retrieval • 9B • Updated • 36.9k • 9
Haon-Chen/speed-embedding-7b-instruct
Feature Extraction • 7B • Updated • 266 • 5
Haon-Chen/speed-synthesis-8b-revisor
Text Generation • 8B • Updated • 4
Haon-Chen/speed-synthesis-8b-senior
Text Generation • 8B • Updated • 8 •