10 citations · 26 across the 18 of their papers we have counts for
21 papers
Robust Decentralized Fairness Auditing
Sayan Biswas, Jade Garcia Bourrée, Anne-Marie Kermarrec +2
Emerging legislation requires large language models (LLMs) to be audited for compliance with regulatory standards, particularly fairness. Such black-box audits typically assume a s…
Backdoor Mitigation in Decentralized LLM Fine-Tuning
Sayan Biswas, Jade Garcia Bourrée, Rachid Guerraoui +5
Decentralized large language model (LLM) fine-tuning lets organizations collaboratively train a shared LLM on data they cannot pool, without a central coordinator. In every round,…
A Unified Benchmark for Privacy-preserving Vector Search
Anne-Marie Kermarrec, Rafael Pires, Mathis Randl +1
Vector search powers semantic search, recommendation systems, and retrieval-augmented generation (RAG). By design, the service answering a query sees both the query embedding and,…
Catapults to the Rescue: Accelerating Vector Search by Exploiting Query Locality
Sami Abuzakuk, Anne-Marie Kermarrec, Rafael Pires +2
Graph-based indexing is the dominant approach for approximate nearest neighbor search in vector databases, offering high recall with low latency across billions of vectors. However…
Mosaic Learning: A Framework for Decentralized Learning with Model Fragmentation
Sayan Biswas, Davide Frey, Romaric Gaudel +7
Decentralized learning (DL) enables collaborative machine learning (ML) without a central server, making it suitable for settings where training data cannot be centrally hosted. We…
Effective LoRA Adapter Routing using Task Representations
Akash Dhasade, Anne-Marie Kermarrec, Igor Pavlovic +4
Low-rank adaptation (LoRA) enables parameter efficient specialization of large language models (LLMs) through modular adapters, resulting in rapidly growing public adapter pools sp…
Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption
Sayan Biswas, Philippe Chartier, Akash Dhasade +7
In contemporary cloud-based services, protecting users' sensitive data and ensuring the confidentiality of the server's model are critical. Fully homomorphic encryption (FHE) enabl…
HarMoEny: Efficient Multi-GPU Inference of MoE Models
Zachary Doucet, Rishi Sharma, Martijn de Vos +3
Mixture-of-Experts (MoE) models offer computational efficiency during inference by activating only a subset of specialized experts for a given input. This enables efficient model s…
Robust ML Auditing using Prior Knowledge
Jade Garcia Bourrée, Augustin Godinot, Martijn De Vos +5
Among the many technical challenges to enforcing AI regulations, one crucial yet underexplored problem is the risk of audit manipulation. This manipulation occurs when a platform d…
Accelerating MoE Model Inference with Expert Sharding
Oana Balmau, Anne-Marie Kermarrec, Rafael Pires +3
Mixture of experts (MoE) models achieve state-of-the-art results in language modeling but suffer from inefficient hardware utilization due to imbalanced token routing and communica…
Practical Federated Learning without a Server
Akash Dhasade, Anne-Marie Kermarrec, Erick Lavoie +3
Federated Learning (FL) enables end-user devices to collaboratively train ML models without sharing raw data, thereby preserving data privacy. In FL, a central parameter server coo…
Efficient Federated Search for Retrieval-Augmented Generation using Lightweight Routing
Akash Dhasade, Rachid Guerraoui, Anne-Marie Kermarrec +4
Large language models (LLMs) achieve remarkable performance across domains but remain prone to hallucinations and inconsistencies. Retrieval-augmented generation (RAG) mitigates th…
Energy-Aware Decentralized Learning with Intermittent Model Training
Akash Dhasade, Paolo Dini, Elia Guerra +5
Decentralized learning (DL) offers a powerful framework where nodes collaboratively train models without sharing raw data and without the coordination of a central server. In the i…
Noiseless Privacy-Preserving Decentralized Learning
Sayan Biswas, Mathieu Even, Anne-Marie Kermarrec +4
Decentralized learning (DL) enables collaborative learning without a server and without training data leaving the users' devices. However, the models shared in DL can still be used…
Fairness Auditing with Multi-Agent Collaboration
Martijn de Vos, Akash Dhasade, Jade Garcia Bourrée +4
Existing work in fairness auditing assumes that each audit is performed independently. In this paper, we consider multiple agents working together, each auditing the same platform…
Epidemic Learning: Boosting Decentralized Learning with Randomized Communication
Martijn de Vos, Sadegh Farhadkhani, Rachid Guerraoui +3
We present Epidemic Learning (EL), a simple yet powerful decentralized learning (DL) algorithm that leverages changing communication topologies to achieve faster model convergence…
Decentralized Learning Made Practical with Client Sampling
Martijn de Vos, Akash Dhasade, Anne-Marie Kermarrec +3
Decentralized learning (DL) leverages edge devices for collaborative model training while avoiding coordination by a central server. Due to privacy concerns, DL has become an attra…
A Deployment-First Methodology to Mechanism Design and Refinement in Distributed Systems
Martijn de Vos, Georgy Ishmaev, Johan Pouwelse +1
Catalyzed by the popularity of blockchain technology, there has recently been a renewed interest in the design, implementation and evaluation of decentralized systems. Most of thes…
Gromit: Benchmarking the Performance and Scalability of Blockchain Systems
Bulat Nasrulin, Martijn De Vos, Georgy Ishmaev +1
The growing number of implementations of blockchain systems stands in stark contrast with still limited research on a systematic comparison of performance characteristics of these…
Bristle: Decentralized Federated Learning in Byzantine, Non-i.i.d. Environments
Joost Verbraeken, Martijn de Vos, Johan Pouwelse
Federated learning (FL) is a privacy-friendly type of machine learning where devices locally train a model on their private data and typically communicate model updates with a serv…
ASTANA: Practical String Deobfuscation for Android Applications Using Program Slicing
Martijn de Vos, Johan Pouwelse
Software obfuscation is widely used by Android developers to protect the source code of their applications against adversarial reverse-engineering efforts. A specific type of obfus…