TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies
Paper
•
2412.10345
•
Published
•
2
CVPR Demo Track
non-profit
AI & ML interests
CVPR Demo Track @ CVPR 2022
Recent Activity
CVPR's activity
jw2yang
authored
a
paper
4 days ago
jw2yang
authored
a
paper
6 days ago
jw2yang
authored
a
paper
16 days ago
lkeab
authored
a
paper
20 days ago
PKUWilliamYang
authored
a
paper
20 days ago
paulgavrikov
authored
9
papers
20 days ago
Can Biases in ImageNet Models Explain Generalization?
Paper
•
2404.01509
•
Published
Are Vision Language Models Texture or Shape Biased and Can We Steer Them?
Paper
•
2403.09193
•
Published
On the Interplay of Convolutional Padding and Adversarial Robustness
Paper
•
2308.06612
•
Published
An Extended Study of Human-like Behavior under Adversarial Training
Paper
•
2303.12669
•
Published
The Power of Linear Combinations: Learning with Random Convolutions
Paper
•
2301.11360
•
Published
Does Medical Imaging learn different Convolution Filters?
Paper
•
2210.13799
•
Published
Robust Models are less Over-Confident
Paper
•
2210.05938
•
Published
Adversarial Robustness through the Lens of Convolutional Filters
Paper
•
2204.02481
•
Published
CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters
Paper
•
2203.15331
•
Published
Post
1747
Qwen/QwQ-32B-Preview shows us the future (and it's going to be exciting)...
I tested it against some really challenging reasoning prompts and the results are amazing 🤯.
Check this dataset for the results: victor/qwq-misguided-attention
I tested it against some really challenging reasoning prompts and the results are amazing 🤯.
Check this dataset for the results: victor/qwq-misguided-attention
- 2 replies
noamrot
authored
a
paper
25 days ago
Post
2220
Perfect example of why
Qwen/Qwen2.5-Coder-32B-Instruct is insane?
Introducing: AI Video Composer 🔥
huggingface-projects/ai-video-composer
Drag and drop your assets (images/videos/audios) to create any video you want using natural language!
It works by asking the model to output a valid FFMPEG and this can be quite complex but most of the time Qwen2.5-Coder-32B gets it right (that thing is a beast). It's an update of an old project made with GPT4 and it was almost impossible to make it work with open models back then (~1.5 years ago), but not anymore, let's go open weights 🚀.
Introducing: AI Video Composer 🔥
huggingface-projects/ai-video-composer
Drag and drop your assets (images/videos/audios) to create any video you want using natural language!
It works by asking the model to output a valid FFMPEG and this can be quite complex but most of the time Qwen2.5-Coder-32B gets it right (that thing is a beast). It's an update of an old project made with GPT4 and it was almost impossible to make it work with open models back then (~1.5 years ago), but not anymore, let's go open weights 🚀.
Post
1820
Qwen2.5-72B is now the default HuggingChat model.
This model is so good that you must try it! I often get better results on rephrasing with it than Sonnet or GPT-4!!
This model is so good that you must try it! I often get better results on rephrasing with it than Sonnet or GPT-4!!
IAMJB
authored
2
papers
about 1 month ago