Foodvision Extended
š„
The extended version of Foodvision mini
Arpit
vapit
Ā·
AI & ML interests
ML Engineer and Data Scientist who loves working with Computer Vision and NLP. Appreciator of Hugging Face's Transformer library. š±āš
Recent Activity
updated a Space about 1 month ago
vapit/foodvision_extended updated a Space 6 months ago
vapit/breemmary updated a Space 6 months ago
vapit/whattocooktodayOrganizations
None yet
updated a
Space about 1 month ago
updated 2
Spaces 6 months ago
updated a
model 11 months ago
published a
model 11 months ago
updated a
model 11 months ago
published a
model 11 months ago
published a
Space 11 months ago
reactedto shekkizh's post with š 12 months ago
Post
1914
Think AGI is just around the corner? Not so fast.
When OpenAI released its Computer-Using Agent (CUA) API, I happened to be playing Wordle š§© and thought, why not see how the model handles it?
Spoiler: Wordle turned out to be a surprisingly effective benchmark.
So Romain Cosentino Ph.D. and I dug in and analyzed the results of several hundred runs.
š Takeaways
1ļøā£ Even the best computer-using models struggle with simple, context-dependent tasks.Ā
2ļøā£ Visual perception and reasoning remain major hurdles for multimodal agents.
3ļøā£ Real-world use cases reveal significant gaps between hype and reality. Perception accuracy drops to near zero by the last turn š
š Read our arxiv article for more details https://www.arxiv.org/abs/2504.15434
When OpenAI released its Computer-Using Agent (CUA) API, I happened to be playing Wordle š§© and thought, why not see how the model handles it?
Spoiler: Wordle turned out to be a surprisingly effective benchmark.
So Romain Cosentino Ph.D. and I dug in and analyzed the results of several hundred runs.
š Takeaways
1ļøā£ Even the best computer-using models struggle with simple, context-dependent tasks.Ā
2ļøā£ Visual perception and reasoning remain major hurdles for multimodal agents.
3ļøā£ Real-world use cases reveal significant gaps between hype and reality. Perception accuracy drops to near zero by the last turn š
š Read our arxiv article for more details https://www.arxiv.org/abs/2504.15434
- 3 replies
updated a
model about 1 year ago
reactedto etemiz's post with š about 1 year ago
Post
2841
My 1 year of work summarized.
TLDR: by carefully curating datasets we can fix misinformation in AI. Then we can use that to measure misinformation in other AI.
https://huggingface.co/blog/etemiz/building-a-beneficial-ai
TLDR: by carefully curating datasets we can fix misinformation in AI. Then we can use that to measure misinformation in other AI.
https://huggingface.co/blog/etemiz/building-a-beneficial-ai
published a
model about 1 year ago
[bot] Conversion to Parquet
#1 opened about 1 year ago
by 
parquet-converter
updated a
dataset about 1 year ago
published a
dataset about 1 year ago
published a
Space about 1 year ago
updated a
model about 1 year ago
published a
model about 1 year ago
updated 2
models about 1 year ago