Loading
Life sciences · Preprint
arXiv · September 8, 2026
Raises a question worth testing. It does not answer one.
This preprint presents a theoretical lattice-based attack on secure aggregation in decentralized federated learning, showing that sparse network topologies can expose private model updates to colluding semi-honest nodes through asymmetric aggregate views. The work establishes a formal connection to the Hidden Subset Sum Problem and demonstrates reconstruction of private states across image, tabular, and text tasks, but lacks peer review and does not evaluate deployed systems or compare against existing defenses.
Theoretical security analysis with proof-of-concept evaluation. Intervention: Lattice-based reconstruction attack combining lattice reduction with structural filtering to recover private model updates from asymmetric aggregate views in sparse decentralized federated learning topologies..
Sparse decentralized topologies create structural leakage: colluding semi-honest nodes obtain asymmetric aggregate views exposing multiple hidden linear combinations of honest participants' private states. Lattice-based reconstruction approach combines lattice reduction with structural filtering to recover protected model states from hidden aggregate views. Attack evaluated on image, tabular, and text tasks; results show colluding nodes can recover original local updates of honest nodes, enabling downstream reconstruction of private training data.
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
The source did not state who this applies to in practice.
A theoretical cryptographic attack demonstrating a privacy vulnerability in decentralized federated learning under specific conditions; proposes a novel attack method but does not evaluate real-world deployments or defenses.
As stated by the source record.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
What is missing. This record has no reported figures. That is a gap in the analysis, not a judgement about the study.
Secure Aggregation (SA) is widely regarded as a strong defense against model-update leakage in Federated Learning (FL), as it reveals only aggregate results while hiding individual updates. In Decentralized Federated Learning (DFL), SA is commonly instantiated as local neighborhood aggregation, where each node obtains a weighted aggregate over its neighbors. We show that this locality creates a structural leakage surface: sparse decentralized topologies provide colluding semi-honest nodes with asymmetric aggregate views, exposing multiple hidden linear combinations of honest participants' private states. Reconstructing private states from these aggregate views is fundamentally challenging, as both the private states and the aggregation coefficients are hidden. We tackle this challenge by establishing a formal connection to the Hidden Subset Sum Problem, a long-studied problem in cryptography. Building on this formulation, we design a lattice-based reconstruction approach that combines lattice reduction with structural filtering to reconstruct protected model states. We evaluate our attack on image, tabular, and text tasks under sparse DFL topologies. Our results show that colluding semi-honest nodes can recover the original local updates of honest nodes, enabling downstream reconstruction of private training data. These findings demonstrate that SA alone does not guarantee privacy in DFL when local aggregation induces asymmetric observations.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.