OmAgent: A Multi-modal Agent Framework for Complex Video Understanding with Task Divide-and-Conquer
Paper
•
2406.16620
•
Published
•
2
zhanglu
zhanglu
·
AI & ML interests
None yet
Recent Activity
authored
a paper
21 days ago
OmAgent: A Multi-modal Agent Framework for Complex Video Understanding
with Task Divide-and-Conquer
liked
a model
7 months ago
lmms-lab/llama3-llava-next-8b
liked
a model
7 months ago
tennant/llava-llama-3-8b-hqedit
Organizations
None yet
zhanglu's activity
authored
a
paper
21 days ago
liked
3
models
7 months ago
reacted to
visheratin's
post with ❤️
8 months ago
Post
Keep stacking cool stuff and getting better results! After I changed the standard vision encoder to SigLIP, NLLB-CLIP got a 10% average performance improvement. And now, I added matryoshka layers (https://arxiv.org/abs/2205.13147) to enable smaller embeddings and got another 6% performance boost! Plus, thanks to MRL, 4.5x smaller embeddings retain 90%+ quality.
The large model is finally SoTA for both image and text multilingual retrieval!
The models are available on the hub:
- visheratin/nllb-siglip-mrl-base
- visheratin/nllb-siglip-mrl-large
The large model is finally SoTA for both image and text multilingual retrieval!
The models are available on the hub:
- visheratin/nllb-siglip-mrl-base
- visheratin/nllb-siglip-mrl-large
- 2 replies
liked
2
Spaces
9 months ago
liked
a
model
11 months ago
liked
a
Space
about 1 year ago
updated
a
collection
about 1 year ago
liked
7
models
about 1 year ago
llmrails/ember-v1
Feature Extraction
•
Updated
•
80.3k
•
61
thenlper/gte-large
Sentence Similarity
•
Updated
•
524k
•
•
266
BAAI/bge-base-en-v1.5
Feature Extraction
•
Updated
•
1.61M
•
•
275
BAAI/bge-large-en-v1.5
Feature Extraction
•
Updated
•
2.23M
•
•
486
WhereIsAI/UAE-Large-V1
Feature Extraction
•
Updated
•
1.14M
•
•
217
intfloat/e5-mistral-7b-instruct
Feature Extraction
•
Updated
•
198k
•
487
meta-llama/Llama-2-7b-chat-hf
Text Generation
•
Updated
•
1.31M
•
•
4.21k
liked
2
models
over 1 year ago
