Hugging Face's logo
Orr Zohar's picture
29 248 28

Orr Zohar PRO

orrzohar
witcher23's profile picture dyyoung's profile picture vinhnx90's profile picture
·
https://orrzohar.github.io

AI & ML interests

Large Multi-Modal Models, Foundation Models, Video Understanding

Recent Activity

upvoted a paper about 21 hours ago
Transformers without Normalization
upvoted a paper 2 days ago
Long Context Tuning for Video Generation
upvoted a paper 2 days ago
VisualWebInstruct: Scaling up Multimodal Instruction Data through Web Search
View all activity

Organizations

Stanford AI's profile picture Hugging Face TB Research's profile picture Hugging Face FineVideo's profile picture smol-explorers's profile picture Apollo-LMMs's profile picture Orr and associates org's profile picture

orrzohar's activity

New activity in HuggingFaceTB/SmolVLM2-500M-Video-Instruct 14 days ago

Update README.md
#15 opened 14 days ago by
orrzohar
New activity in HuggingFaceTB/SmolVLM2-256M-Video-Instruct 14 days ago

Update README.md
#8 opened 14 days ago by
orrzohar
New activity in HuggingFaceTB/SmolVLM2-2.2B-Instruct 14 days ago

Input Video length constraints

8
#6 opened 23 days ago by
NikhilJoson
New activity in HuggingFaceTB/SmolVLM2-2.2B-Instruct 18 days ago

update tag

4
#11 opened 18 days ago by
orrzohar
New activity in HuggingFaceTB/SmolVLM2-2.2B-Instruct 19 days ago

Several questions on the same video

1
#8 opened 22 days ago by
delphijb
New activity in HuggingFaceTB/SmolVLM2-2.2B-Instruct 23 days ago

checkpoint you are trying to load has model type `smolvlm` but Transformers does not recognize this

1
#7 opened 23 days ago by
JLouisBiz
New activity in HuggingFaceTB/SmolVLM2-500M-Video-Instruct 23 days ago

update readme dtype
#8 opened 23 days ago by
orrzohar
New activity in HuggingFaceTB/SmolVLM2-256M-Video-Instruct 23 days ago

update readme dtype
#5 opened 23 days ago by
orrzohar
New activity in HuggingFaceTB/SmolVLM2-2.2B-Instruct 23 days ago

Fix dtype processing in README example
#5 opened 23 days ago by
orrzohar

RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (CUDABFloat16Type) should be the same

2
#4 opened 23 days ago by
Neiko2002

Using pre-computed embeddings for images/frames and using as input

3
#2 opened 24 days ago by
maxlun
New activity in HuggingFaceTB/SmolVLM2-500M-Video-Instruct 28 days ago

README update
#5 opened 28 days ago by
orrzohar
New activity in nvidia/Eagle2-9B about 2 months ago

Sequence packing logic

1
#3 opened about 2 months ago by
orrzohar
commented 2 papers 3 months ago
New activity in lmms-lab/LLaVA-OneVision-Data 6 months ago

Missing/corrupted images in dataset
#9 opened 6 months ago by
orrzohar
commented a paper 7 months ago
New activity in ShareGPT4Video/ShareGPT4Video 7 months ago

obtaining original videos for video instruction tuning
#23 opened 7 months ago by
orrzohar
commented 2 papers 8 months ago