11 citations · 25 across the 10 of their papers we have counts for
12 papers
Safety Reconstructed: Generative Modeling via Masked Diffusion Builds Strong Safety Guardrails
Gert Lek, Abele Malan, Chaoyi Zhu +3
Guard models are the last line of defense between a language model and a harmful output, yet their training objective is surprisingly narrow. Existing guards learn to predict a sin…
Distributed Edge Inference: an Experimental Study on Multiview Detection
Gianluca Mittone, Giulio Malenza, Marco Aldinucci +1
Computing is evolving rapidly to cater to the increasing demand for sophisticated services, and Cloud computing lays a solid foundation for flexible on-demand provisioning. However…
Efficiently Distributed Federated Learning
Gianluca Mittone, Robert Birke, Marco Aldinucci
Federated Learning (FL) is experiencing a substantial research interest, with many frameworks being developed to allow practitioners to build federations easily and quickly. Most o…
TMPDiff: Temporal Mixed-Precision for Diffusion Models
Basile Lewandowski, Simon Kurz, Aditya Shankar +3
Diffusion models are the go-to method for Text-to-Image generation, but their iterative denoising processes has high inference latency. Quantization reduces compute time by using l…
Match & Choose: Model Selection Framework for Fine-tuning Text-to-Image Diffusion Models
Basile Lewandowski, Robert Birke, Lydia Y. Chen
Text-to-image (T2I) models based on diffusion and transformer architectures advance rapidly. They are often pretrained on large corpora, and openly shared on a model platform, such…
Building an Accelerated OpenFOAM Proof-of-Concept Application using Modern C++
Giulio Malenza, Giovanni Stabile, Filippo Spiga +2
The modern trend in High-Performance Computing (HPC) involves the use of accelerators such as Graphics Processing Units (GPUs) alongside Central Processing Units (CPUs) to speed up…
Optimization-Free Universal Watermark Forgery with Regenerative Diffusion Models
Chaoyi Zhu, Zaitang Li, Renyi Yang +4
Watermarking becomes one of the pivotal solutions to trace and verify the origin of synthetic images generated by artificial intelligence models, but it is not free of risks. Recen…
Exploring energy consumption of AI frameworks on a 64-core RV64 Server CPU
Giulio Malenza, Francesco Targa, Adriano Marques Garcia +2
In today's era of rapid technological advancement, artificial intelligence (AI) applications require large-scale, high-performance, and data-intensive computations, leading to sign…
SFDDM: Single-fold Distillation for Diffusion models
Chi Hong, Jiyue Huang, Robert Birke +3
While diffusion models effectively generate remarkable synthetic images, a key limitation is the inference inefficiency, requiring numerous sampling steps. To accelerate inference…
DALLMi: Domain Adaption for LLM-based Multi-label Classifier
Miruna Beţianu, Abele Mălan, Marco Aldinucci +2
Large language models (LLMs) increasingly serve as the backbone for classifying text associated with distinct domains and simultaneously several labels (classes). When encountering…
TabuLa: Harnessing Language Models for Tabular Data Synthesis
Zilong Zhao, Robert Birke, Lydia Chen
Tabular data synthesis is crucial for addressing privacy and security concerns in industries reliant on tabular data. While recent advancements adopt large language models (LLMs) f…
Multi-Label Gold Asymmetric Loss Correction with Single-Label Regulators
Cosmin Octavian Pene, Amirmasoud Ghiassi, Taraneh Younesian +2
Multi-label learning is an emerging extension of the multi-class classification where an image contains multiple labels. Not only acquiring a clean and fully labeled dataset in mul…