EdgeDeID de-identifies clinical notes locally without real patient training data
On a 100-note synthetic benchmark it reaches 0.994 micro-F1, but F1 falls to 0.48 on MIMIC ICU nursing notes.
Stories per day · 30 days
2 RSS
Relatedaudio-visual1benchmark1clinical-notes1de-identification1privacy1speaker-extraction1speech-processing1synthetic-data1
7 stories, newest first · one card per story
Stories tagged #edge-inference
Falcon cuts causal AV speaker extraction cost on new RealSSA benchmark
Falcon reports 91.9% fewer parameters and 11.1 times lower GPU latency than causal AV-TFGridNet while topping RealSSA-Sim extraction quality.
Building AI for Reliable Execution: Lessons From Industrial Robotics
Standard Bots raised $200 million at a $1 billion valuation to scale AI-native industrial robot arms.
9
Oído: Open-Vocabulary Speech Recognition on a $5 Microcontroller without a Neural Accelerator
Oído runs open-vocabulary English speech recognition on a $5 ESP32-S3 without a neural accelerator.
ESP32S3 cluster running 1.58-bit (BitNet) Language model
A GitHub project runs a 1.58-bit BitNet language model on an ESP32-S3 cluster.
MicroQonv: Reshaping Convolution Tensors for Efficient Microscaling in Training and Inference
MicroQonv reduces convolutional microscaling cost by quantizing tensors once before a modified im2col.
The Other Half of the Memory Wall: Serving 35B MoEs from SSD with Trained Routing Prediction
Edge0 streams 35B-parameter MoE inference from SSD on a 24GB machine at 20 tokens/s using a one-token-ahead prerouter, with framework and checkpoints open-sourced.