dinhanhx's picture

dinhanhx

dinhanhx

·

dinhanhx

AI & ML interests

Vision Language

Recent Activity

replied to merve's post 1 day ago

Apollo is a new family of open-source video language models by Meta, where 3B model outperforms most 7B models and 7B outperforms most 30B models 🧶 ✨ the models come in 1.5B https://huggingface.co/Apollo-LMMs/Apollo-1_5B-t32, 3B https://huggingface.co/Apollo-LMMs/Apollo-3B-t32 and 7B https://huggingface.co/Apollo-LMMs/Apollo-7B-t32 with A2.0 license, based on Qwen1.5 & Qwen2 ✨ the authors also release a benchmark dataset https://huggingface.co/spaces/Apollo-LMMs/ApolloBench The paper has a lot of experiments (they trained 84 models!) about what makes the video LMs work ⏯️ Try the demo for best setup here https://huggingface.co/spaces/Apollo-LMMs/Apollo-3B they evaluate sampling strategies, scaling laws for models and datasets, video representation and more! > The authors find out that whatever design decision was applied to small models also scale properly when the model and dataset are scaled 📈 scaling dataset has diminishing returns for smaller models > They evaluate frame sampling strategies, and find that FPS sampling is better than uniform sampling, and they find 8-32 tokens per frame optimal > They also compare image encoders, they try a variation of models from shape optimized SigLIP to DINOv2 they find https://huggingface.co/google/siglip-so400m-patch14-384 to be most powerful 🔥 > they also compare freezing different parts of models, training all stages with some frozen parts give the best yield They eventually release three models, where Apollo-3B outperforms most 7B models and Apollo 7B outperforms 30B models 🔥

liked a model 1 day ago

erax-ai/EraX-VL-2B-V1.5

liked a model 8 days ago

HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1.5

View all activity

Organizations

dinhanhx's activity

New activity in CohereForAI/c4ai-command-r7b-12-2024 18 days ago

ValueError when use Vllm

#5 opened 19 days ago by

New activity in dinhanhx/velvet 8 months ago

Apply for community grant: Academic project (gpu and storage)

#1 opened 12 months ago by

New activity in statmt/cc100 about 1 year ago

did not find Arabic dataset

#2 opened almost 2 years ago by

Create language_list.tsv

#3 opened about 1 year ago by

New activity in oscar-corpus/OSCAR-2109 about 1 year ago

Example of downloading a part of this repos

#3 opened about 1 year ago by

New activity in jbochi/madlad400-8b-lm about 1 year ago

Vocab size of tokenizer is not equal to vocab size of the model

#2 opened about 1 year ago by

New activity in YoungMasterFromSect/Trauter_LoRAs about 1 year ago

Some questions about the Lora training Process

#29 opened almost 2 years ago by

New activity in facebook/convnextv2-huge-22k-512 over 1 year ago

Image_size should be 512

#1 opened over 1 year ago by

New activity in vinai/phobert-base over 2 years ago

Change to RobertaModel

#1 opened over 2 years ago by