Publications

From Efficiency to Equity: Measuring Fairness in Preference Learning

S. Gowaikar

Rashid A. Mushkani

Shin Koseki

As AI systems, particularly generative models, increasingly influence decision-making, ensuring that they are able to fairly represent diver… (voir plus)se human preferences becomes crucial. This paper introduces a novel framework for evaluating epistemic fairness in preference learning models inspired by economic theories of inequality and Rawlsian justice. We propose metrics adapted from the Gini Coefficient, Atkinson Index, and Kuznets Ratio to quantify fairness in these models. We validate our approach using two datasets: a custom visual preference dataset (AI-EDI-Space) and the Jester Jokes dataset. Our analysis reveals variations in model performance across users, highlighting potential epistemic injustices. We explore pre-processing and in-processing techniques to mitigate these inequalities, demonstrating a complex relationship between model efficiency and fairness. This work contributes to AI ethics by providing a framework for evaluating and improving epistemic fairness in preference learning models, offering insights for developing more inclusive AI systems in contexts where diverse human preferences are crucial.

2025-10-14

Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (publié)

doi.org

arxiv.org

International AI Safety Report 2025: First Key Update: Capabilities and Risk Implications

Yoshua Bengio

Stephen Clare

Carina Prunkl

Maksym Andriushchenko

BEN BUCKNALL

Philip Fox

Tiancheng Hu

Cameron Jones

Sam Manning

Nestor Maslej

Vasilios Mavroudis

Conor McGlynn

Malcolm Murray

Shalaleh Rismani

Charlotte Stix

Lucia Velasco

Nicole Wheeler

Daniel Privitera

Sören Mindermann

Daron Acemoglu … (voir 36 de plus)

Thomas G. Dietterich

Fredrik Heintz

Geoffrey Hinton

Nick Jennings

Susan Leavy

Teresa Ludermir

Vidushi Marda

Helen Margetts

John McDermid

Jane Munga

Arvind Narayanan

Alondra Nelson

Clara Neppel

Sarvapali D. (Gopal) Ramchurn

Stuart Russell

Marietje Schaake

Bernhard Schölkopf

Alvaro Soto

Lee Tiedrich

Gael Varoquaux

Andrew Yao

Ya-Qin Zhang

Lambrini Das

Claire Dennis

Arianna Dini

Freya Hempleman

Samuel Kenny

Patrick King

Hannah Merchant

Jamie-Day Rawal

Rose Woolhouse

Jonathan Barry

Marc-Antoine Guérard

Claire Latendresse

Cassidy MacNeil

Benjamin Prud’homme

Since the publication of the first International AI Safety Report, AI capabilities have continued to improve across key domains. New trainin… (voir plus)g techniques that teach AI systems to reason step-by-step and inference-time enhancements have primarily driven these advances, rather than simply training larger models. As a result, general-purpose AI systems can solve more complex problems in a range of domains, from scientific research to software development. Their performance on benchmarks that measure performance in coding, mathematics, and answering expert-level science questions has continued to improve, though reliability challenges persist, with systems excelling on some tasks while failing completely on others. These capability improvements also have implications for multiple risks, including risks from biological weapons and cyber attacks. Finally, they pose new challenges for monitoring and controllability. This update examines how AI capabilities have improved since the first Report, then focuses on key risk areas where substantial new evidence warrants updated assessments.

2025-10-14

arXiv (prépublication)

doi.org

arxiv.org

Longitudinal intergenerational hyperscanning indexes changes in social connection

Ryssa Moffat

Guillaume Dumas

Emily S. Cross

Loneliness is globally acknowledged as a severe and burgeoning health risk, fuelling interest in helping people of all ages form meaningful … (voir plus)social connections. One promising approach consists of intergenerational social programs. While behavioural and qualitative evidence derived from such programs promise health and wellbeing benefits, the physiological consequences of repeated intergenerational encounters remain unknown. Insight into physiological changes will shed light on the mechanisms of social connection and can inform program design choices. We charted changes in interpersonal neural synchrony (INS) in 31 intergenerational (older/younger adult) and 30 same generation (younger adult) dyads across a six-session creative drawing program. At each session, dyads completed self-report measures, drew together and alone, and had their cortical activation recorded with fNIRS. In both groups, INS was greater while dyads drew together than alone. Across sessions, intergenerational dyads’ INS decreased and same generation dyads’ INS increased. INS in RIFG∼RTPJ and RIFG∼RIFG were predictive of loneliness levels and feelings of social closeness, respectively. The research reinforces the multi-faceted nature of INS dynamics as social connections are forged.

2025-10-14

bioRxiv (prépublication)

doi.org

Measuring What Matters: Connecting AI Ethics Evaluations to System Attributes, Hazards, and Harms

Renee Shelby

Over the past decade, an ecosystem of measures has emerged to evaluate the social and ethical implications of AI systems, largely shaped by … (voir plus)high-level ethics principles. These measures are developed and used in fragmented ways, without adequate attention to how they are situated in AI systems. In this paper, we examine how existing measures used in the computing literature map to AI system components, attributes, hazards, and harms. Our analysis draws on a scoping review resulting in nearly 800 measures corresponding to 11 AI ethics principles. We find that most measures focus on four principles – fairness, transparency, privacy, and trust – and primarily assess model or output system components. Few measures account for interactions across system elements, and only a narrow set of hazards is typically considered for each harm type. Many measures are disconnected from where harm is experienced and lack guidance for setting meaningful thresholds. These patterns reveal how current evaluation practices remain fragmented, measuring in pieces rather than capturing how harms emerge across systems. Framing measures with respect to system attributes, hazards, and harms can strengthen regulatory oversight, support actionable practices in industry, and ground future research in systems-level understanding.

2025-10-14

Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (publié)

doi.org

arxiv.org

PoissonNet: A Local-Global Approach for Learning on Surfaces

Arman Maesumi

Tanish Makadia

Thibault Groueix

Vladimir Kim

Daniel Ritchie

Noam Aigerman

Many network architectures exist for learning on meshes, yet their constructions entail delicate trade-offs between difficulty learning high… (voir plus)-frequency features, insufficient receptive field, sensitivity to discretization, and inefficient computational overhead. Drawing from classic local-global approaches in mesh processing, we introduce PoissonNet, a novel neural architecture that overcomes all of these deficiencies by formulating a local-global learning scheme, which uses Poisson's equation as the primary mechanism for feature propagation. Our core network block is simple; we apply learned local feature transformations in the gradient domain of the mesh, then solve a Poisson system to propagate scalar feature updates across the surface globally. Our local-global learning framework preserves the features's full frequency spectrum and provides a truly global receptive field, while remaining agnostic to mesh triangulation. Our construction is efficient, requiring far less compute overhead than comparable methods, which enables scalability -- both in the size of our datasets, and the size of individual training samples. These qualities are validated on various experiments where, compared to previous intrinsic architectures, we attain state-of-the-art performance on semantic segmentation and parameterizing highly-detailed animated surfaces. Finally, as a central application of PoissonNet, we show its ability to learn deformations, significantly outperforming state-of-the-art architectures that learn on surfaces.

2025-10-14

ArXiv (prépublication)

doi.org

arxiv.org

Reframing AI-for-Good: Radical Questioning in AI for Human Trafficking Interventions

Gabriel Lefebvre

This paper introduces Radical Questioning (RQ), a structured, pre-design ethics framework developed to assess whether artificial intelligenc… (voir plus)e (AI) should be applied to complex social problems rather than merely how. While much of responsible AI development focuses on aligning systems with principles such as fairness, transparency, and accountability, it often begins after the decision to build has already been made, implicitly treating the deployment of AI as a given rather than a question in itself. In domains such as human trafficking, marked by contested definitions, systemic injustice, and deep stakeholder asymmetries, such assumptions can obscure foundational ethical concerns. RQ offers an upstream, deliberative process for surfacing these concerns before design begins. Drawing from critical theory, participatory ethics, and relational responsibility, RQ formalizes a five-step framework to interrogate problem framings, confront techno-solutionist tendencies, and reflect on the moral legitimacy of intervention. Developed through interdisciplinary collaboration and engagement with survivor-led organizations, RQ was piloted in the domain of human trafficking (HT) which is a particularly high-stakes and ethically entangled application area. Its use led to a fundamental design shift: away from automated detection tools and toward survivor-controlled, empowerment-based technologies. We argue that RQ's novelty lies in both its temporal position, i.e, prior to technical design, and its orientation toward domains where harm is structural and ethical clarity cannot be achieved through one-size-fits-all solutions. RQ thus addresses a critical gap between abstract principles of responsible AI and the lived ethical demands of real-world deployment.

2025-10-14

AAAI/ACM Conference on AI, Ethics, and Society (publié)

doi.org

Simplicial Embeddings Improve Sample Efficiency in Actor-Critic Agents

Johan Samir Obando Ceron

Samuel Lavoie

Recent works have proposed accelerating the wall-clock training time of actor-critic methods via the use of large-scale environment parallel… (voir plus)ization; unfortunately, these can sometimes still require large number of environment interactions to achieve a desired level of performance. Noting that well-structured representations can improve the generalization and sample efficiency of deep reinforcement learning (RL) agents, we propose the use of simplicial embeddings: lightweight representation layers that constrain embeddings to simplicial structures. This geometric inductive bias results in sparse and discrete features that stabilize critic bootstrapping and strengthen policy gradients. When applied to FastTD3, FastSAC, and PPO, simplicial embeddings consistently improve sample efficiency and final performance across a variety of continuous- and discrete-control environments, without any loss in runtime speed.

2025-10-14

ArXiv (prépublication)

doi.org

arxiv.org

Systemizing Multiplicity: The Curious Case of Arbitrariness in Machine Learning

Prakhar Ganesh

Afaf Taïk

Golnoosh Farnadi

Algorithmic modeling relies on limited information in data to extrapolate outcomes for unseen scenarios, often embedding an element of arbit… (voir plus)rariness in its decisions. A perspective on this arbitrariness that has recently gained interest is multiplicity-the study of arbitrariness across a set of "good models", i.e., those likely to be deployed in practice. In this work, we systemize the literature on multiplicity by: (a) formalizing the terminology around model design choices and their contribution to arbitrariness, (b) expanding the definition of multiplicity to incorporate underrepresented forms beyond just predictions and explanations, (c) clarifying the distinction between multiplicity and other lenses of arbitrariness, i.e., uncertainty and variance, and (d) distilling the benefits and potential risks of multiplicity into overarching trends, situating it within the broader landscape of responsible AI. We conclude by identifying open research questions and highlighting emerging trends in this young but rapidly growing area of research.

2025-10-14

AAAI/ACM Conference on AI Ethics and Society (publié)

doi.org

arxiv.org

The Interpolation Constraint in the RV Analysis of M-Dwarfs Using Empirical Templates

Dhvani Doshi

Nicolas B. Cowan

E. Artigau

René Doyon

André M. Silva

Khaled Al Moulla

Yashar Hezaveh

Precise radial velocity (pRV) measurements of M-dwarfs in the near-infrared (NIR) rely on empirical templates due to the lack of accurate st… (voir plus)ellar spectral models in this regime. Templates are assumed to approximate the true spectrum when constructed from many observations or in the high signal-to-noise limit. We develop a numerical simulation that generates SPIRou-like pRV observations from PHOENIX spectra, constructs empirical templates, and estimates radial velocities. This simulation solely considers photon noise and evaluates when empirical templates remain reliable for pRV analysis. Our results reveal a previously unrecognized noise source in templates, establishing a fundamental floor for template-based pRV measurements. We find that templates inherently include distortions in stellar line shapes due to imperfect interpolation at the detector's sampling resolution. The magnitude of this interpolation error depends on sampling resolution and RV content. Consequently, while stars with a higher RV content, such as cooler M-dwarfs are expected to yield lower RV uncertainties, their dense spectral features can amplify interpolation errors, potentially biasing RV estimates. For a typical M4V star, SPIRou's spectral and sampling resolution imposes an RV uncertainty floor of 0.5-0.8 m/s, independent of the star's magnitude or the telescope's aperture. These findings reveal a limitation of template-based pRV methods, underscoring the need for improved spectral modeling and better-than-Nyquist detector sampling to reach the next level of RV precision.

2025-10-14

The Astronomical Journal (publié)

doi.org

arxiv.org

Joint Satellite Power Consumption and Handover Optimization for LEO Constellations

Yassine Afif

Mohammed Almekhlafi

Antoine Lesage-Landry

Gunes Karabulut Kurt

In satellite constellation-based communication systems, continuous user coverage requires frequent handoffs due to the dynamic topology indu… (voir plus)ced by the Low Earth Orbit (LEO) satellites. Each handoff between a satellite and ground users introduces additional signaling and power consumption, which can become a significant burden as the size of the constellation continues to increase. This work focuses on the optimization of the total transmission rate in a LEO-to-user system, by jointly considering the total transmitted power, user-satellite associations, and power consumption, the latter being handled through a penalty on handoff events. We consider a system where LEO satellites serve users located in remote areas with no terrestrial connectivity, and formulate the power allocation problem as a mixed-integer concave linear program (MICP) subject to power and association constraints. Our approach can be solved with off-the-shelf solvers and is benchmarked against a naive baseline where users associate to their closest visible satellite. Extensive Monte Carlo simulations demonstrate the effectiveness of the proposed method in controlling the handoff frequency while maintaining high user throughput. These performance gains highlight the effectiveness of our handover-aware optimization strategy, which ensures that user rates improve significantly, by about 40%, without incurring a disproportionate rise in the handoff frequency.

2025-10-13

International Conference on Wireless for Space and Extreme Environments (publié)

doi.org

arxiv.org

Personalized Federated Fine-Tuning of Vision Foundation Models for Healthcare

Adam Tupper

Christian Gagné

2025-10-13

ArXiv (prépublication)

doi.org

arxiv.org

RNAGenScape: Property-guided Optimization and Interpolation of mRNA Sequences with Manifold Langevin Dynamics