iil-postech/semantic-attention
Official implementation of "Attention-aware semantic communications for collaborative inference” (IEEE IoTJ 2024)
This project helps machine learning engineers and researchers optimize collaborative inference systems by reducing communication costs. It takes an image dataset and a trained vision model, then applies attention-aware patch selection and uncertainty measures to determine which parts of the image and which images are most crucial. The output is a more efficient collaboration between edge devices and central servers, maintaining high classification accuracy while minimizing data transfer.
Use this if you are developing or deploying distributed computer vision systems where edge devices and central servers collaborate, and you need to reduce the amount of data transmitted without significantly impacting accuracy.
Not ideal if your system is not bottlenecked by communication costs between devices, or if you are not working with image classification tasks using attention-based models.
Stars
13
Forks
3
Language
Jupyter Notebook
License
—
Category
Last pushed
Jan 22, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/iil-postech/semantic-attention"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
lucidrains/x-transformers
A concise but complete full-attention transformer with a set of promising experimental features...
kanishkamisra/minicons
Utility for behavioral and representational analyses of Language Models
lucidrains/simple-hierarchical-transformer
Experiments around a simple idea for inducing multiple hierarchical predictive model within a GPT
lucidrains/dreamer4
Implementation of Danijar's latest iteration for his Dreamer line of work
Nicolepcx/Transformers-in-Action
This is the corresponding code for the book Transformers in Action