Google’s next-gen federated learning uses TEEs for verifiable, auditable anonymization—logic can be remotely attested while staying confidential/integrity-protected. Faster Gboard compute too. 🔒🧠⚙️ #FederatedLearning #DifferentialPrivacy #TrustedExecutionEnvironments #MachineLearning #Privacy
https://research.google/blog/toward-provably-private-learning-from-federated-data/
research.google
November 6, 2026 at 4:39 PM
Google’s next-gen federated learning uses TEEs for verifiable, auditable anonymization—logic can be remotely attested while staying confidential/integrity-protected. Faster Gboard compute too. 🔒🧠⚙️ #FederatedLearning #DifferentialPrivacy #TrustedExecutionEnvironments #MachineLearning #Privacy
The Digital Health & Discovery Platform (#DHDP) is using #federatedlearning to help researchers analyze health data across institutions without moving patient records.
#precisionmedicine
#precisionmedicine
October 30, 2026 at 6:55 PM
The Digital Health & Discovery Platform (#DHDP) is using #federatedlearning to help researchers analyze health data across institutions without moving patient records.
#precisionmedicine
#precisionmedicine
Canada's health data could help drive new discoveries and more personalized care. But sharing sensitive patient information comes with real privacy challenges.
www.dhdp.ca/updates/news...
#CyberMonth2026
#DigitalHealth #HealthResearch #FederatedLearning
www.dhdp.ca/updates/news...
#CyberMonth2026
#DigitalHealth #HealthResearch #FederatedLearning
October 30, 2026 at 6:55 PM
Canada's health data could help drive new discoveries and more personalized care. But sharing sensitive patient information comes with real privacy challenges.
www.dhdp.ca/updates/news...
#CyberMonth2026
#DigitalHealth #HealthResearch #FederatedLearning
www.dhdp.ca/updates/news...
#CyberMonth2026
#DigitalHealth #HealthResearch #FederatedLearning
🎓 Two guest talks at BME, open to all!
🔹 Dr. Hao Wang: Robust & personalized federated learning
🔹 Dr. Miao Pan: Neural audio watermark security & attacks
📅 Oct 28, 4:30 PM
📍 BME Building I, IB023, Budapest
#AISecurity #CrySyS #FederatedLearning
🔹 Dr. Hao Wang: Robust & personalized federated learning
🔹 Dr. Miao Pan: Neural audio watermark security & attacks
📅 Oct 28, 4:30 PM
📍 BME Building I, IB023, Budapest
#AISecurity #CrySyS #FederatedLearning
October 26, 2026 at 8:05 AM
🎓 Two guest talks at BME, open to all!
🔹 Dr. Hao Wang: Robust & personalized federated learning
🔹 Dr. Miao Pan: Neural audio watermark security & attacks
📅 Oct 28, 4:30 PM
📍 BME Building I, IB023, Budapest
#AISecurity #CrySyS #FederatedLearning
🔹 Dr. Hao Wang: Robust & personalized federated learning
🔹 Dr. Miao Pan: Neural audio watermark security & attacks
📅 Oct 28, 4:30 PM
📍 BME Building I, IB023, Budapest
#AISecurity #CrySyS #FederatedLearning
A new framework called AdaFedXplain combines adaptive federated learning with explainable AI to improve ransomware detection while addressing data privacy, communication cost, and non-IID data challenges.
#Semiconductors #AIInfrastructure #FederatedLearning
https://arxiv.org/abs/2610.25318
#Semiconductors #AIInfrastructure #FederatedLearning
https://arxiv.org/abs/2610.25318
October 20, 2026 at 6:01 PM
A new framework called AdaFedXplain combines adaptive federated learning with explainable AI to improve ransomware detection while addressing data privacy, communication cost, and non-IID data challenges.
#Semiconductors #AIInfrastructure #FederatedLearning
https://arxiv.org/abs/2610.25318
#Semiconductors #AIInfrastructure #FederatedLearning
https://arxiv.org/abs/2610.25318
A new paper introduces FederatedBytes, a tokenizer-free, byte-level small language model with linear attention designed to lower the computational and communication costs of federated learning for NLP and…
#FederatedLearning #MultimodalAI #NLP #EdgeComputing
https://arxiv.org/abs/2610.22906
#FederatedLearning #MultimodalAI #NLP #EdgeComputing
https://arxiv.org/abs/2610.22906
October 19, 2026 at 4:01 PM
A new paper introduces FederatedBytes, a tokenizer-free, byte-level small language model with linear attention designed to lower the computational and communication costs of federated learning for NLP and…
#FederatedLearning #MultimodalAI #NLP #EdgeComputing
https://arxiv.org/abs/2610.22906
#FederatedLearning #MultimodalAI #NLP #EdgeComputing
https://arxiv.org/abs/2610.22906
A new paper proposes FedSkillGit, a federated skill evolution approach for LLM agents that avoids repeatedly uploading nearly unchanged full skill documents. It aims to address privacy leakage, wasted…
#AI #FederatedLearning #LLMAgents #DataInfrastructure
https://arxiv.org/abs/2610.21348
#AI #FederatedLearning #LLMAgents #DataInfrastructure
https://arxiv.org/abs/2610.21348
October 17, 2026 at 4:01 PM
A new paper proposes FedSkillGit, a federated skill evolution approach for LLM agents that avoids repeatedly uploading nearly unchanged full skill documents. It aims to address privacy leakage, wasted…
#AI #FederatedLearning #LLMAgents #DataInfrastructure
https://arxiv.org/abs/2610.21348
#AI #FederatedLearning #LLMAgents #DataInfrastructure
https://arxiv.org/abs/2610.21348
A new arXiv paper, Optimal Hydra, proposes a data-informed approach to model-heterogeneous federated learning for LLMs, tailoring sub-networks to client capabilities and data distributions rather than relying…
#FederatedLearning #LLM #ModelCompression #AIResearch
https://arxiv.org/abs/2610.20753
#FederatedLearning #LLM #ModelCompression #AIResearch
https://arxiv.org/abs/2610.20753
October 17, 2026 at 2:01 PM
A new arXiv paper, Optimal Hydra, proposes a data-informed approach to model-heterogeneous federated learning for LLMs, tailoring sub-networks to client capabilities and data distributions rather than relying…
#FederatedLearning #LLM #ModelCompression #AIResearch
https://arxiv.org/abs/2610.20753
#FederatedLearning #LLM #ModelCompression #AIResearch
https://arxiv.org/abs/2610.20753
New arXiv paper proposes AirPASS, a framework combining over-the-air federated learning with pinching antenna systems to jointly optimize device selection, beamforming, and antenna placement. Targets scalable…
#FederatedLearning #WirelessAI #PinchingAntenna #EdgeAI
https://arxiv.org/abs/2607.06768
#FederatedLearning #WirelessAI #PinchingAntenna #EdgeAI
https://arxiv.org/abs/2607.06768
October 17, 2026 at 12:01 AM
New arXiv paper proposes AirPASS, a framework combining over-the-air federated learning with pinching antenna systems to jointly optimize device selection, beamforming, and antenna placement. Targets scalable…
#FederatedLearning #WirelessAI #PinchingAntenna #EdgeAI
https://arxiv.org/abs/2607.06768
#FederatedLearning #WirelessAI #PinchingAntenna #EdgeAI
https://arxiv.org/abs/2607.06768
FedSumUp adds a lossy compressor to federated data condensation, giving users compact representations they can audit and reducing privacy leakage on resource-constrained devices. Server-trained and…
#FedSumUp #FederatedLearning #PrivacyAware #EdgeAI
https://arxiv.org/abs/2610.21194
#FedSumUp #FederatedLearning #PrivacyAware #EdgeAI
https://arxiv.org/abs/2610.21194
October 16, 2026 at 12:01 PM
FedSumUp adds a lossy compressor to federated data condensation, giving users compact representations they can audit and reducing privacy leakage on resource-constrained devices. Server-trained and…
#FedSumUp #FederatedLearning #PrivacyAware #EdgeAI
https://arxiv.org/abs/2610.21194
#FedSumUp #FederatedLearning #PrivacyAware #EdgeAI
https://arxiv.org/abs/2610.21194
SHADE enables quantum models to forget data without re-measuring quantum states. Using classical shadows recorded during training, federated nodes estimate gradients and apply Newton corrections to remove forget data in <0.2s, matching retraining accuracy.
#QuantumML #FederatedLearning #Research
#QuantumML #FederatedLearning #Research
Classical Shadow-Based Unlearning for Quantum Federated Learning
arxiv.org
October 15, 2026 at 6:02 AM
SHADE enables quantum models to forget data without re-measuring quantum states. Using classical shadows recorded during training, federated nodes estimate gradients and apply Newton corrections to remove forget data in <0.2s, matching retraining accuracy.
#QuantumML #FederatedLearning #Research
#QuantumML #FederatedLearning #Research
Federated learning hides who contributed, not what the model memorizes. New ACL 2026 work: FL models memorize client data - your own prefixes pull your own text back out best - with no clear reduction vs centralized training. Update privacy isn't model privacy. #federatedlearning #privacy #ML
October 15, 2026 at 12:15 PM
Federated learning hides who contributed, not what the model memorizes. New ACL 2026 work: FL models memorize client data - your own prefixes pull your own text back out best - with no clear reduction vs centralized training. Update privacy isn't model privacy. #federatedlearning #privacy #ML
#KER
A comprehensive survey maps poisoning attacks against federated learning, from Byzantine gradients to backdoor triggers, and benchmarks robust defenses under non-IID data conditions.
Details: https://www.maxapress.com/article/doi/10.48130/ker-0026-0014
#FederatedLearning #PoisoningAttacks
A comprehensive survey maps poisoning attacks against federated learning, from Byzantine gradients to backdoor triggers, and benchmarks robust defenses under non-IID data conditions.
Details: https://www.maxapress.com/article/doi/10.48130/ker-0026-0014
#FederatedLearning #PoisoningAttacks
October 15, 2026 at 12:34 AM
#KER
A comprehensive survey maps poisoning attacks against federated learning, from Byzantine gradients to backdoor triggers, and benchmarks robust defenses under non-IID data conditions.
Details: https://www.maxapress.com/article/doi/10.48130/ker-0026-0014
#FederatedLearning #PoisoningAttacks
A comprehensive survey maps poisoning attacks against federated learning, from Byzantine gradients to backdoor triggers, and benchmarks robust defenses under non-IID data conditions.
Details: https://www.maxapress.com/article/doi/10.48130/ker-0026-0014
#FederatedLearning #PoisoningAttacks
Your local RL round trains on your transcripts alone. Everyone's encrypted updates aggregate into one global - tomorrow's adapter knows work no single user ever did. Your model file is never touched. https://github.com/roger-federated/roger-federated #federatedlearning #selfhosted #FOSS
October 13, 2026 at 12:17 PM
Your local RL round trains on your transcripts alone. Everyone's encrypted updates aggregate into one global - tomorrow's adapter knows work no single user ever did. Your model file is never touched. https://github.com/roger-federated/roger-federated #federatedlearning #selfhosted #FOSS
A new paper proposes Personalized Federated Direct Preference Optimization (pFedDPO), letting each client keep a small local LoRA adapter on a shared base model to better fit heterogeneous…
#FederatedLearning #LoRA #PreferenceOptimization #PrivacyPreservingML
https://arxiv.org/abs/2610.12885
#FederatedLearning #LoRA #PreferenceOptimization #PrivacyPreservingML
https://arxiv.org/abs/2610.12885
October 12, 2026 at 6:01 PM
A new paper proposes Personalized Federated Direct Preference Optimization (pFedDPO), letting each client keep a small local LoRA adapter on a shared base model to better fit heterogeneous…
#FederatedLearning #LoRA #PreferenceOptimization #PrivacyPreservingML
https://arxiv.org/abs/2610.12885
#FederatedLearning #LoRA #PreferenceOptimization #PrivacyPreservingML
https://arxiv.org/abs/2610.12885
Fed-GRPO proposes a federated extension of Group Relative Policy Optimization for LLM fine-tuning, using GRPO's own reward statistics as zero-cost signals to drive aggregation, local training, and communication.…
#FederatedLearning #LLM #RLHF #PrivacyPreservingML
https://arxiv.org/abs/2610.11502
#FederatedLearning #LLM #RLHF #PrivacyPreservingML
https://arxiv.org/abs/2610.11502
October 10, 2026 at 2:01 PM
Fed-GRPO proposes a federated extension of Group Relative Policy Optimization for LLM fine-tuning, using GRPO's own reward statistics as zero-cost signals to drive aggregation, local training, and communication.…
#FederatedLearning #LLM #RLHF #PrivacyPreservingML
https://arxiv.org/abs/2610.11502
#FederatedLearning #LLM #RLHF #PrivacyPreservingML
https://arxiv.org/abs/2610.11502
Federated Learning on Stochastic Neural Networks
www.dl.begellhouse.com/journals/558...
#FederatedLearning #NeuralNetworks #MachineLearning
www.dl.begellhouse.com/journals/558...
#FederatedLearning #NeuralNetworks #MachineLearning
October 6, 2026 at 2:45 PM
Federated Learning on Stochastic Neural Networks
www.dl.begellhouse.com/journals/558...
#FederatedLearning #NeuralNetworks #MachineLearning
www.dl.begellhouse.com/journals/558...
#FederatedLearning #NeuralNetworks #MachineLearning
A new pre-deployment framework estimates task difficulty for federated perception systems by combining intrinsic data properties with client-distribution composition. Experiments on three MNIST variants show strong negative…
#EdgeAI #FederatedLearning #AIResearch
https://arxiv.org/abs/2603.28282
#EdgeAI #FederatedLearning #AIResearch
https://arxiv.org/abs/2603.28282
October 6, 2026 at 6:01 PM
A new pre-deployment framework estimates task difficulty for federated perception systems by combining intrinsic data properties with client-distribution composition. Experiments on three MNIST variants show strong negative…
#EdgeAI #FederatedLearning #AIResearch
https://arxiv.org/abs/2603.28282
#EdgeAI #FederatedLearning #AIResearch
https://arxiv.org/abs/2603.28282
Federated learning's pitch: fine-tune locally, share only tiny LoRA gradients. MineGrad (arXiv 2608.01521) breaks it - a malicious server hides extraction in the pretrained model and reads your training data out of them. The server is the attack surface. #privacy #federatedlearning #ML
October 5, 2026 at 12:21 PM
Federated learning's pitch: fine-tune locally, share only tiny LoRA gradients. MineGrad (arXiv 2608.01521) breaks it - a malicious server hides extraction in the pretrained model and reads your training data out of them. The server is the attack surface. #privacy #federatedlearning #ML
Stacking several LoRA deltas onto one base? Summing them scales the effective step with N - each delta was trained to move the weights a given distance, and the sum overshoots every one of them. Average them (1/N) instead. #LoRA #MachineLearning #federatedlearning
October 2, 2026 at 12:21 PM
Stacking several LoRA deltas onto one base? Summing them scales the effective step with N - each delta was trained to move the weights a given distance, and the sum overshoots every one of them. Average them (1/N) instead. #LoRA #MachineLearning #federatedlearning
This arXiv paper proposes vFedProtoQNAS, a federated learning approach where clients train device-specific quantum neural networks and share class prototypes instead of model parameters. It targets…
#QuantumAI #FederatedLearning #EdgeAI #NeuralArchitectureSearch
https://arxiv.org/abs/2610.01718
#QuantumAI #FederatedLearning #EdgeAI #NeuralArchitectureSearch
https://arxiv.org/abs/2610.01718
October 2, 2026 at 4:01 PM
This arXiv paper proposes vFedProtoQNAS, a federated learning approach where clients train device-specific quantum neural networks and share class prototypes instead of model parameters. It targets…
#QuantumAI #FederatedLearning #EdgeAI #NeuralArchitectureSearch
https://arxiv.org/abs/2610.01718
#QuantumAI #FederatedLearning #EdgeAI #NeuralArchitectureSearch
https://arxiv.org/abs/2610.01718
Fixed: roger resolved a fork like gemma-4-12B-it-heretic to its base by name containment - so its chats would have trained the base federation's global. It resolves exactly now. #federatedlearning #opensource #selfhosted https://github.com/roger-federated/roger-federated
October 1, 2026 at 12:42 PM
Fixed: roger resolved a fork like gemma-4-12B-it-heretic to its base by name containment - so its chats would have trained the base federation's global. It resolves exactly now. #federatedlearning #opensource #selfhosted https://github.com/roger-federated/roger-federated
New federated learning approach HO-FL splits each model into a zeroth-order bottom segment and a first-order top segment, letting edge devices pick the boundary based on their memory budget while training one…
#FederatedLearning #EdgeAI #ModelTraining #AIResearch
https://arxiv.org/abs/2609.39074
#FederatedLearning #EdgeAI #ModelTraining #AIResearch
https://arxiv.org/abs/2609.39074
October 1, 2026 at 2:01 PM
New federated learning approach HO-FL splits each model into a zeroth-order bottom segment and a first-order top segment, letting edge devices pick the boundary based on their memory budget while training one…
#FederatedLearning #EdgeAI #ModelTraining #AIResearch
https://arxiv.org/abs/2609.39074
#FederatedLearning #EdgeAI #ModelTraining #AIResearch
https://arxiv.org/abs/2609.39074
A new arXiv study, "Unapologetically Distributed," provides the first broad look at distributed learning in Document Analysis across tasks, architectures, and fine-tuning strategies. It challenges the…
#FederatedLearning #DocumentAnalysis #PrivacyPreserving #AI
https://arxiv.org/abs/2609.39684
#FederatedLearning #DocumentAnalysis #PrivacyPreserving #AI
https://arxiv.org/abs/2609.39684
October 1, 2026 at 4:01 PM
A new arXiv study, "Unapologetically Distributed," provides the first broad look at distributed learning in Document Analysis across tasks, architectures, and fine-tuning strategies. It challenges the…
#FederatedLearning #DocumentAnalysis #PrivacyPreserving #AI
https://arxiv.org/abs/2609.39684
#FederatedLearning #DocumentAnalysis #PrivacyPreserving #AI
https://arxiv.org/abs/2609.39684
Manchester Metropolitan University is offering a PhD Studentship in Lightweight Trust-Aware Private Federated Learning for Secure UAV Swarms under Model Poisoning Attacks.
mlscientist.com/phd-students...
#FederatedLearning
mlscientist.com/phd-students...
#FederatedLearning
September 30, 2026 at 6:06 PM
Manchester Metropolitan University is offering a PhD Studentship in Lightweight Trust-Aware Private Federated Learning for Secure UAV Swarms under Model Poisoning Attacks.
mlscientist.com/phd-students...
#FederatedLearning
mlscientist.com/phd-students...
#FederatedLearning