48 citations · 139 across the 37 of their papers we have counts for
44 papers
Mise-en-Scène: Implicit Layout Emergence in Diffusion Transformers for Human-AI Design Co-Creation
Zipeng Xu, Ryan Murdock, Umberto Michieli
Automating graphic design synthesis from user-provided elements requires both a coherent overall composition and the exact preservation of each asset. Existing methods predict a la…
Feature-Space Generative Models for One-Shot Class-Incremental Learning
Jack Foster, Kirill Paramonov, Mete Ozay +1
Few-shot class-incremental learning (FSCIL) is a paradigm where a model, initially trained on a dataset of base classes, must adapt to an expanding problem space by recognizing nov…
Clustering-driven Memory Compression for On-device Large Language Models
Ondrej Bohdal, Pramit Saha, Umberto Michieli +2
Large language models (LLMs) often rely on user-specific memories distilled from past interactions to enable personalized generation. A common practice is to concatenate these memo…
Data-driven Clustering and Merging of Adapters for On-device Large Language Models
Ondrej Bohdal, Taha Ceritli, Mete Ozay +4
On-device large language models commonly employ task-specific adapters (e.g., LoRAs) to deliver strong performance on downstream tasks. While storing all available adapters is impr…
MemLoRA: Distilling Expert Adapters for On-Device Memory Systems
Massimo Bini, Ondrej Bohdal, Umberto Michieli +3
Memory-augmented Large Language Models (LLMs) have demonstrated remarkable consistency during prolonged dialogues by storing relevant memories and incorporating them as context. Su…
CG-TTRL: Context-Guided Test-Time Reinforcement Learning for On-Device Large Language Models
Peyman Hosseini, Ondrej Bohdal, Taha Ceritli +4
Test-time Reinforcement Learning (TTRL) has shown promise in adapting foundation models for complex tasks at test-time, resulting in large performance improvements. TTRL leverages…
On-device System of Compositional Multi-tasking in Large Language Models
Ondrej Bohdal, Konstantinos Theodosiadis, Asterios Mpatziakas +10
Large language models (LLMs) are commonly adapted for diverse downstream tasks via parameter-efficient fine-tuning techniques such as Low-Rank Adapters (LoRA). While adapters can b…
K-Merge: Online Continual Merging of Adapters for On-device Large Language Models
Donald Shenaj, Ondrej Bohdal, Taha Ceritli +3
On-device deployment of Large Language Models (LLMs) frequently leverages Low-Rank Adapters (LoRAs) to support diverse downstream tasks under tight resource constraints. To address…
FedPromo: Federated Lightweight Proxy Models at the Edge Bring New Domains to Foundation Models
Matteo Caligiuri, Francesco Barbato, Donald Shenaj +2
Federated Learning (FL) is an established paradigm for training deep learning models on decentralized data. However, as the size of the models grows, conventional FL approaches oft…
HydraOpt: Navigating the Efficiency-Performance Trade-off of Adapter Merging
Taha Ceritli, Ondrej Bohdal, Mete Ozay +4
Large language models (LLMs) often leverage adapters, such as low-rank-based adapters, to achieve strong performance on downstream tasks. However, storing a separate adapter for ea…
Efficient Compositional Multi-tasking for On-device Large Language Models
Ondrej Bohdal, Mete Ozay, Jijoong Moon +3
Adapter parameters provide a mechanism to modify the behavior of machine learning models and have gained significant popularity in the context of large language models (LLMs) and g…
Controllable Forgetting Mechanism for Few-Shot Class-Incremental Learning
Kirill Paramonov, Mete Ozay, Eunju Yang +2
Class-incremental learning in the context of limited personal labeled samples (few-shot) is critical for numerous real-world applications, such as smart home devices. A key challen…
LoRA.rar: Learning to Merge LoRAs via Hypernetworks for Subject-Style Conditioned Image Generation
Donald Shenaj, Ondrej Bohdal, Mete Ozay +2
Recent advancements in image generation models have enabled personalized image creation with both user-defined subjects (content) and styles. Prior works achieved personalization b…
DreamCache: Finetuning-Free Lightweight Personalized Image Generation via Feature Caching
Emanuele Aiello, Umberto Michieli, Diego Valsesia +2
Personalized image generation requires text-to-image generative models that capture the core features of a reference subject to allow for controlled generation across different con…
Randomized Asymmetric Chain of LoRA: The First Meaningful Theoretical Framework for Low-Rank Adaptation
Grigory Malinovsky, Umberto Michieli, Hasan Abed Al Kader Hammoud +4
Fine-tuning has become a popular approach to adapting large foundational models to specific tasks. As the size of models and datasets grows, parameter-efficient fine-tuning techniq…
Swiss DINO: Efficient and Versatile Vision Framework for On-device Personal Object Search
Kirill Paramonov, Jia-Xing Zhong, Umberto Michieli +2
In this paper, we address a recent trend in robotic home appliances to include vision systems on personal devices, capable of personalizing the appliances on the fly. In particular…
Cross-Architecture Auxiliary Feature Space Translation for Efficient Few-Shot Personalized Object Detection
Francesco Barbato, Umberto Michieli, Jijoong Moon +2
Recent years have seen object detection robotic systems deployed in several personal devices (e.g., home robots and appliances). This has highlighted a challenge in their design, i…
Model Merging and Safety Alignment: One Bad Model Spoils the Bunch
Hasan Abed Al Kader Hammoud, Umberto Michieli, Fabio Pizzati +4
Merging Large Language Models (LLMs) is a cost-effective technique for combining multiple expert LLMs into a single versatile model, retaining the expertise of the original ones. H…
Object-conditioned Bag of Instances for Few-Shot Personalized Instance Recognition
Umberto Michieli, Jijoong Moon, Daehyun Kim +1
Nowadays, users demand for increased personalization of vision systems to localize and identify personal instances of objects (e.g., my dog rather than dog) from a few-shot dataset…
FFT-based Selection and Optimization of Statistics for Robust Recognition of Severely Corrupted Images
Elena Camuffo, Umberto Michieli, Jijoong Moon +2
Improving model robustness in case of corrupted images is among the key challenges to enable robust vision systems on smart devices, such as robotic agents. Particularly, robust te…
A Modular System for Enhanced Robustness of Multimedia Understanding Networks via Deep Parametric Estimation
Francesco Barbato, Umberto Michieli, Mehmet Kerim Yucel +2
In multimedia understanding tasks, corrupted samples pose a critical challenge, because when fed to machine learning models they lead to performance degradation. In the past, three…
Deep Neural Network Models Trained With A Fixed Random Classifier Transfer Better Across Domains
Hafiz Tiomoko Ali, Umberto Michieli, Ji Joong Moon +2
The recently discovered Neural collapse (NC) phenomenon states that the last-layer weights of Deep Neural Networks (DNN), converge to the so-called Equiangular Tight Frame (ETF) si…
HOP to the Next Tasks and Domains for Continual Learning in NLP
Umberto Michieli, Mete Ozay
Continual Learning (CL) aims to learn a sequence of problems (i.e., tasks and domains) by transferring knowledge acquired on previous problems, whilst avoiding forgetting of past o…
RECALL+: Adversarial Web-based Replay for Continual Learning in Semantic Segmentation
Chang Liu, Giulia Rizzoli, Francesco Barbato +5
Catastrophic forgetting of previous knowledge is a critical issue in continual learning typically handled through various regularization strategies. However, existing methods strug…