Picture for Francesco Conti

Francesco Conti

VEXP: A Low-Cost RISC-V ISA Extension for Accelerated Softmax Computation in Transformers

Add code
Apr 15, 2025
Viaarxiv icon

Lightweight Software Kernels and Hardware Extensions for Efficient Sparse Deep Neural Networks on Microcontrollers

Add code
Mar 08, 2025
Viaarxiv icon

Open-Source Heterogeneous SoCs for AI: The PULP Platform Experience

Add code
Dec 29, 2024
Figure 1 for Open-Source Heterogeneous SoCs for AI: The PULP Platform Experience
Figure 2 for Open-Source Heterogeneous SoCs for AI: The PULP Platform Experience
Figure 3 for Open-Source Heterogeneous SoCs for AI: The PULP Platform Experience
Figure 4 for Open-Source Heterogeneous SoCs for AI: The PULP Platform Experience
Viaarxiv icon

Accelerating Image-based Pest Detection on a Heterogeneous Multi-core Microcontroller

Add code
Aug 29, 2024
Viaarxiv icon

Deeploy: Enabling Energy-Efficient Deployment of Small Language Models On Heterogeneous Microcontrollers

Add code
Aug 08, 2024
Figure 1 for Deeploy: Enabling Energy-Efficient Deployment of Small Language Models On Heterogeneous Microcontrollers
Figure 2 for Deeploy: Enabling Energy-Efficient Deployment of Small Language Models On Heterogeneous Microcontrollers
Figure 3 for Deeploy: Enabling Energy-Efficient Deployment of Small Language Models On Heterogeneous Microcontrollers
Figure 4 for Deeploy: Enabling Energy-Efficient Deployment of Small Language Models On Heterogeneous Microcontrollers
Viaarxiv icon

Toward Attention-based TinyML: A Heterogeneous Accelerated Architecture and Automated Deployment Flow

Add code
Aug 05, 2024
Viaarxiv icon

Distilling Tiny and Ultra-fast Deep Neural Networks for Autonomous Navigation on Nano-UAVs

Add code
Jul 17, 2024
Viaarxiv icon

Multi-resolution Rescored ByteTrack for Video Object Detection on Ultra-low-power Embedded Systems

Add code
Apr 17, 2024
Viaarxiv icon

Optimizing the Deployment of Tiny Transformers on Low-Power MCUs

Add code
Apr 03, 2024
Figure 1 for Optimizing the Deployment of Tiny Transformers on Low-Power MCUs
Figure 2 for Optimizing the Deployment of Tiny Transformers on Low-Power MCUs
Figure 3 for Optimizing the Deployment of Tiny Transformers on Low-Power MCUs
Figure 4 for Optimizing the Deployment of Tiny Transformers on Low-Power MCUs
Viaarxiv icon

Combining Local and Global Perception for Autonomous Navigation on Nano-UAVs

Add code
Mar 18, 2024
Figure 1 for Combining Local and Global Perception for Autonomous Navigation on Nano-UAVs
Figure 2 for Combining Local and Global Perception for Autonomous Navigation on Nano-UAVs
Figure 3 for Combining Local and Global Perception for Autonomous Navigation on Nano-UAVs
Viaarxiv icon