Tootfinder

Per-element Secure Aggregation against Data Reconstruction Attacks in Federated Learning
Takumi Suimon, Yuki Koizumi, Junji Takemasa, Toru Hasegawa
https://arxiv.org/abs/2508.04285

Per-element Secure Aggregation against Data Reconstruction Attacks in Federated Learning
Federated learning (FL) enables collaborative model training without sharing raw data, but individual model updates may still leak sensitive information. Secure aggregation (SecAgg) mitigates this risk by allowing the server to access only the sum of client updates, thereby concealing individual contributions. However, a significant vulnerability has recently attracted increasing attention: when model updates are sparse vectors, a non-zero value contributed by a single client at a given index c…

On the Normalization of Confusion Matrices: Methods and Geometric Interpretations
Johan Erbani, Pierre-Edouard Portier, Elod Egyed-Zsigmond, Sonia Ben Mokhtar, Diana Nurbakova
https://arxiv.org/abs/2509.04959

On the Normalization of Confusion Matrices: Methods and Geometric Interpretations
The confusion matrix is a standard tool for evaluating classifiers by providing insights into class-level errors. In heterogeneous settings, its values are shaped by two main factors: class similarity -- how easily the model confuses two classes -- and distribution bias, arising from skewed distributions in the training and test sets. However, confusion matrix values reflect a mix of both factors, making it difficult to disentangle their individual contributions. To address this, we introduce b…

Non-Linear Counterfactual Aggregate Optimization
Benjamin Heymann, Otmane Sakhi
https://arxiv.org/abs/2509.03438 https://arxiv.org/pdf/2509.03438

Non-Linear Counterfactual Aggregate Optimization
We consider the problem of directly optimizing a non-linear function of an outcome, where this outcome itself is the sum of many small contributions. The non-linearity of the function means that the problem is not equivalent to the maximization of the expectation of the individual contribution. By leveraging the concentration properties of the sum of individual outcomes, we derive a scalable descent algorithm that directly optimizes for our stated objective. This allows for instance to maximize…

Electron-positron scattering at finite temperature in Podolsky electrodynamics
D. S. Cabral, L. A. S. Evangelista, L. H. A. R. Ferreira, A. F. Santos
https://arxiv.org/abs/2509.06568

Electron-positron scattering at finite temperature in Podolsky electrodynamics
The electron-positron annihilation process is investigated within the framework of Podolsky's generalized electrodynamics at finite temperature. In this theory, a higher-derivative term modifies the photonic kinetic sector, introducing a massive mode while preserving gauge invariance. Thermal effects are incorporated using the real-time Thermo Field Dynamics formalism. The total cross section is calculated, and the individual contributions of the Podolsky parameter and thermal effects are analy…

Collective contributions to polarization in political voting
Gavin Rees, Edward D. Lee
https://arxiv.org/abs/2508.02496 https://arxiv.org/pdf/2508.02496

Collective contributions to polarization in political voting
Politics around the world exhibits increasing polarization, demonstrated in part by rigid voting configurations in legislatures. The crux of polarization is separation along a unidimensional ideological axis, but how it emerges is yet partially understood. We refine a powerful class of models from statistical physics, restricted Boltzmann machines, to unify two classes of individual voter preference and voter interaction models. Our model is minimally parameterized, fits voting data well, and h…

A comprehensive separation of dark matter and baryonic mass components in galaxy clusters I: Mass constraints from Abell S1063
Benjamin Beauchesne, Benjamin Cl\'ement, Marceau Limousin, Bel\'en Alcalde Pampliega, Mathilde Jauzac, Anna Niemiec, Johan Richard, Guillaume Mahler, Jose M. Diego, Pascale Hibon, Anton M. Koekemoer, Thomas Connor, Jean-Paul Kneib, Andreas L. Faisst

A comprehensive separation of dark matter and baryonic mass components in galaxy clusters I: Mass constraints from Abell S1063
In this two-part series, we present a multi-probe mass modelling method for massive galaxy clusters, designed to disentangle the contributions of individual mass components (Dark matter, intra-cluster gas, stellar masses). In this first paper, we focus on recovering the mass constraint datasets required for the modelling approach introduced in the second paper. Specifically, we measure the light distribution, stellar mass, and kinematics of the cluster members, the brightest cluster galaxy (BCG…

Authorship-contribution normalized Sh-index and citations are better research output indicators
Vishvesh Karthik, Indupalli Sishir Anand, Utkarsha Mahanta, Gaurav Sharma
https://arxiv.org/abs/2509.04124

Authorship-contribution normalized Sh-index and citations are better research output indicators
Bibliometric measures, such as total citations and h-index, have become a cornerstone for evaluating academic performance; however, these traditional metrics, being non-weighted, inadequately capture the nuances of individual contributions. To address this constraint, we developed GScholarLens, an open-access browser extension that integrates seamlessly with Google Scholar to enable detailed bibliometric analysis. GScholarLens categorizes publications by authorship roles, adjusts citation weigh…

Understanding Action Effects through Instrumental Empowerment in Multi-Agent Reinforcement Learning
Ardian Selmonaj, Miroslav Strupl, Oleg Szehr, Alessandro Antonucci
https://arxiv.org/abs/2508.15652

Understanding Action Effects through Instrumental Empowerment in Multi-Agent Reinforcement Learning
To reliably deploy Multi-Agent Reinforcement Learning (MARL) systems, it is crucial to understand individual agent behaviors within a team. While prior work typically evaluates overall team performance based on explicit reward signals or learned value functions, it is unclear how to infer agent contributions in the absence of any value feedback. In this work, we investigate whether meaningful insights into agent behaviors can be extracted that are consistent with the underlying value functions,…

PRAC3 (Privacy, Reputation, Accountability, Consent, Credit, Compensation): Long Tailed Risks of Voice Actors in AI Data-Economy
Tanusree Sharma, Yihao Zhou, Visar Berisha
https://arxiv.org/abs/2507.16247

PRAC3 (Privacy, Reputation, Accountability, Consent, Credit, Compensation): Long Tailed Risks of Voice Actors in AI Data-Economy
Early large-scale audio datasets, such as LibriSpeech, were built with hundreds of individual contributors whose voices were instrumental in the development of speech technologies, including audiobooks and voice assistants. Yet, a decade later, these same contributions have exposed voice actors to a range of risks. While existing ethical frameworks emphasize Consent, Credit, and Compensation (C3), they do not adequately address the emergent risks involving vocal identities that are increasingly…

SpeCT: A state-of-the-art tool to calculate correlated-k tables and continua of CO$_2$-H$_2$O-N$_2$ gas mixtures
G. Chaverot, M. Turbet, H. Tran, J. -M. Hartmann, A. Campargue, D. Mondelain, E. Bolmont
https://arxiv.org/abs/2508.18049

SpeCT: A state-of-the-art tool to calculate correlated-k tables and continua of CO$_2$-H$_2$O-N$_2$ gas mixtures
A key challenge in modeling (exo)planetary atmospheres lies in generating extensive opacity datasets that cover the wide variety of possible atmospheric composition, pressure, and temperature conditions. This critical step requires specific knowledge and can be considerably time-consuming. To circumvent this issue, most available codes approximate the total opacity by summing the contributions of individual molecular species during the radiative transfer calculation. This approach neglects inte…

Detect \& Score: Privacy-Preserving Misbehaviour Detection and Contribution Evaluation in Federated Learning
Marvin Xhemrishi, Alexandre Graell i Amat, Bal\'azs Pej\'o
https://arxiv.org/abs/2506.23583

Detect \& Score: Privacy-Preserving Misbehaviour Detection and Contribution Evaluation in Federated Learning
Federated learning with secure aggregation enables private and collaborative learning from decentralised data without leaking sensitive client information. However, secure aggregation also complicates the detection of malicious client behaviour and the evaluation of individual client contributions to the learning. To address these challenges, QI (Pejo et al.) and FedGT (Xhemrishi et al.) were proposed for contribution evaluation (CE) and misbehaviour detection (MD), respectively. QI, however, l…