João Palmeiro

joaompalmeiro's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

liked a model about 3 hours ago

smolagents/SmolVLM2-2.2B-Instruct-Agentic-GUI

Image-to-Text • 2B • Updated Sep 18, 2025 • 274 • 58

upvoted an article about 3 hours ago

view article

Article

Open Responses: What you need to know

• 
• 
• +2

1 day ago

•

49

liked a Space about 3 hours ago

Running on Zero

Featured

36

Moondream3

🏢

36

Image and video tasks with moondream3.

upvoted a collection about 4 hours ago

YOLO26 Models

Collection

YOLO26 models: detection, segmentation, classification, pose, and OBB variants with demo Space. • 31 items • Updated about 23 hours ago • 9

liked a model about 10 hours ago

kyutai/pocket-tts

Updated 1 day ago • 221

liked 2 models 8 days ago

Qwen/Qwen3-VL-Embedding-2B

Image-to-Text • 2B • Updated 7 days ago • 46.2k • 240

Qwen/Qwen3-VL-Embedding-8B

Image-to-Text • 8B • Updated 7 days ago • 36.5k • 259

reacted to Akhil-Theerthala's post with 👍 9 days ago

view post

Post

405

Is it better to show a model too many images once (Diversity), or extract as much information from a small set of images?

I have always wanted to do an ablation study on this and recently I got the chance to do exactly that. Why? In applied domains like robotics, manufacturing, or banking, we rarely have the luxury of internet-scale diverse image datasets. We are often "Data Poor" in terms of diversity but "Data Rich" in depth.

The takeaway? Density is efficient for facts but dangerous for reasoning (logical collapse) if you don't have larger scale data.

More details:
https://huggingface.co/blog/Akhil-Theerthala/diversity-density-for-vision-language-models

See translation

• 
• 1 reply

·

👍

3

3

+

Reply

upvoted a collection 9 days ago

Density-Vs-Diversity-Blogpost

Collection

The collection contains the artefacts used to do the analysis for the blogpost: Diversity Vs Density: A strategy comparison for fine-tuning VLMs • 7 items • Updated 10 days ago • 2

upvoted an article 10 days ago

view article

Article

Diversity Vs Density: A data strategy comparison for fine-tuning VLMs

• 

11 days ago

•

5

liked a model 24 days ago

jinaai/jina-vlm

Image-Text-to-Text • 2B • Updated Dec 5, 2025 • 3.47k • 94

liked a dataset about 1 month ago

axel-darmouni/anychart-vqa

Viewer • Updated May 9, 2025 • 300 • 7 • 8

liked a model about 1 month ago

google/embeddinggemma-300m

Sentence Similarity • 0.3B • Updated Sep 25, 2025 • 599k • • 1.42k

liked a dataset about 1 month ago

eyehole/VisChainBench

Updated May 20, 2025 • 73 • 2

upvoted a collection about 1 month ago

Ministral 3

Collection

A collection of edge models, with Base, Instruct and Reasoning variants, in 3 different sizes: 3B, 8B and 14B. All with vision capabilities. • 9 items • Updated Dec 2, 2025 • 146

liked a model 2 months ago

nvidia/nemotron-graphic-elements-v1

Object Detection • Updated Dec 11, 2025 • 1 • 14

liked 2 models 3 months ago

utter-project/TowerVision-9B

Image-Text-to-Text • 10B • Updated Nov 5, 2025 • 33 • 4

utter-project/TowerVision-2B

Image-Text-to-Text • 3B • Updated Nov 5, 2025 • 27 • 4

liked a dataset 3 months ago

TIGER-Lab/VisPlotBench

Viewer • Updated Nov 3, 2025 • 888 • 280 • 2

upvoted a collection 3 months ago

Qwen3-VL

Collection

37 items • Updated 16 days ago • 579