Publications

MILUV: A Multi-UAV Indoor Localization dataset with UWB and Vision
Mohammed Ayman Shalaby
Nicholas Dahdah
Syed Shabbir Ahmed
Charles Champagne Cossette
Jerome Le Ny
James Richard Forbes
This paper introduces MILUV, a Multi-UAV Indoor Localization dataset with UWB and Vision measurements. This dataset comprises 217 minutes of… (voir plus) flight time over 36 experiments using three quadcopters, collecting ultra-wideband (UWB) ranging data such as the raw timestamps and channel-impulse response data, vision data from a stereo camera and a bottom-facing monocular camera, inertial measurement unit data, height measurements from a laser rangefinder, magnetometer data, and ground-truth poses from a motion-capture system. The UWB data is collected from up to 12 transceivers affixed to mobile robots and static tripods in both line-of-sight and non-line-of-sight conditions. The UAVs fly at a maximum speed of 4.418 m/s in an indoor environment with visual fiducial markers as features. MILUV is versatile and can be used for a wide range of applications beyond localization, but the primary purpose of MILUV is for testing and validating multi-robot UWB- and vision-based localization algorithms. The dataset can be downloaded at https://doi.org/10.25452/figshare.plus.28386041.v1. A development kit is presented alongside the MILUV dataset, which includes benchmarking algorithms such as visual-inertial odometry, UWB-based localization using an extended Kalman filter, and classification of CIR data using machine learning approaches. The development kit can be found at https://github.com/decargroup/miluv, and is supplemented with a website available at https://decargroup.github.io/miluv/.
2026-01-08
International Journal of Robotics Research (publié)
doi.org
arxiv.org
Multilingual Amnesia: On the Transferability of Unlearning in Multilingual LLMs
Alireza Dehghanpour Farashah
Aditi Khandelwal
Marylou Fauchard
Zhuan Shi
Negar Rostamzadeh
Golnoosh Farnadi
As multilingual large language models become more widely used, ensuring their safety and fairness across diverse linguistic contexts present… (voir plus)s unique challenges. While existing research on machine unlearning has primarily focused on monolingual settings, typically English, multilingual environments introduce additional complexities due to cross-lingual knowledge transfer and biases embedded in both pretraining and fine-tuning data. In this work, we study multilingual unlearning using the Aya-Expanse 8B model under two settings: (1) data unlearning and (2) concept unlearning. We extend benchmarks for factual knowledge and stereotypes to ten languages through translation: English, French, Arabic, Japanese, Russian, Farsi, Korean, Hindi, Hebrew, and Indonesian. These languages span five language families and a wide range of resource levels. Our experiments show that unlearning in high-resource languages is generally more stable, with asymmetric transfer effects observed between typologically related languages. Furthermore, our analysis of linguistic distances indicates that syntactic similarity is the strongest predictor of cross-lingual unlearning behavior.
2026-01-08
arXiv (prépublication)
doi.org
arxiv.org
Tracing Stereotypes in Pre-trained Transformers: From Biased Neurons to Fairer Models
Gianmario Voria
Moses Openja
Foutse Khomh
Gemma Catolino
Fabio Palomba
The advent of transformer-based language models has reshaped how AI systems process and generate text. In software engineering (SE), these m… (voir plus)odels now support diverse activities, accelerating automation and decision-making. Yet, evidence shows that these models can reproduce or amplify social biases, raising fairness concerns. Recent work on neuron editing has shown that internal activations in pre-trained transformers can be traced and modified to alter model behavior. Building on the concept of knowledge neurons, neurons that encode factual information, we hypothesize the existence of biased neurons that capture stereotypical associations within pre-trained transformers. To test this hypothesis, we build a dataset of biased relations, i.e., triplets encoding stereotypes across nine bias types, and adapt neuron attribution strategies to trace and suppress biased neurons in BERT models. We then assess the impact of suppression on SE tasks. Our findings show that biased knowledge is localized within small neuron subsets, and suppressing them substantially reduces bias with minimal performance loss. This demonstrates that bias in transformers can be traced and mitigated at the neuron level, offering an interpretable approach to fairness in SE.
2026-01-08
ArXiv (prépublication)
doi.org
arxiv.org
Large language models can effectively convince people to believe conspiracies
Thomas H Costello
Kellin Pelrine
Matthew Kowal
Antonio A. Arechar
Jean-François Godbout
Adam Gleave
David G. Rand
Gordon Pennycook
Large language models (LLMs) have been shown to be persuasive across a variety of contexts. But it remains unclear whether this persuasive p… (voir plus)ower advantages truth over falsehood, or if LLMs can promote misbeliefs just as easily as refuting them. Here, we investigate this question across three pre-registered experiments in which participants (N = 2,724 Americans) discussed a conspiracy theory they were uncertain about with GPT-4o, and the model was instructed to either argue against ("debunking") or for ("bunking") that conspiracy. When using a"jailbroken"GPT-4o variant with guardrails removed, the AI was as effective at increasing conspiracy belief as decreasing it. Concerningly, the bunking AI was rated more positively, and increased trust in AI, more than the debunking AI. Surprisingly, we found that using standard GPT-4o produced very similar effects, such that the guardrails imposed by OpenAI did little to prevent the LLM from promoting conspiracy beliefs. Encouragingly, however, a corrective conversation reversed these newly induced conspiracy beliefs, and simply prompting GPT-4o to only use accurate information dramatically reduced its ability to increase conspiracy beliefs. Our findings demonstrate that LLMs possess potent abilities to promote both truth and falsehood, but that potential solutions may exist to help mitigate this risk.
2026-01-07
ArXiv (prépublication)
doi.org
arxiv.org
Learning Latent Action World Models In The Wild
Quentin Garrido
Tushar Nagarajan
Basile Terver
Nicolas Ballas
Yann Lecun
Michael G. Rabbat
Agents capable of reasoning and planning in the real world require the ability of predicting the consequences of their actions. While world … (voir plus)models possess this capability, they most often require action labels, that can be complex to obtain at scale. This motivates the learning of latent action models, that can learn an action space from videos alone. Our work addresses the problem of learning latent actions world models on in-the-wild videos, expanding the scope of existing works that focus on simple robotics simulations, video games, or manipulation data. While this allows us to capture richer actions, it also introduces challenges stemming from the video diversity, such as environmental noise, or the lack of a common embodiment across videos. To address some of the challenges, we discuss properties that actions should follow as well as relevant architectural choices and evaluations. We find that continuous, but constrained, latent actions are able to capture the complexity of actions from in-the-wild videos, something that the common vector quantization does not. We for example find that changes in the environment coming from agents, such as humans entering the room, can be transferred across videos. This highlights the capability of learning actions that are specific to in-the-wild videos. In the absence of a common embodiment across videos, we are mainly able to learn latent actions that become localized in space, relative to the camera. Nonetheless, we are able to train a controller that maps known actions to latent ones, allowing us to use latent actions as a universal interface and solve planning tasks with our world model with similar performance as action-conditioned baselines. Our analyses and experiments provide a step towards scaling latent action models to the real world.
2026-01-07
ArXiv (prépublication)
doi.org
arxiv.org
Seeing the forest and the trees: a workflow for automatic acquisition of ultra-high resolution drone photos of tropical forest canopies to support botanical and ecological studies
Étienne Laliberté
Antoine Caron-Guay
Vincent Le Falher
Guillaume Tougas
Helene C. Muller-Landau
Gonzalo Rivas-Torres
Thomas R. Walla
Hugo Baudchon
Mélvin Hernandez
Adrian Buenaño
Anna Weber
Jeffrey Q. Chambers
Jomber Chota Inuma
Fernando Araúz
Jorge Valdes
Andrés Hernández
David Brassfield
P. Sérgio
Vicente Vasquez
Adriana Simonetti … (voir 7 de plus)
Daniel Magnabosco Marra
Caroline de Moura Vasconcelos
Jarol Fernando Vaca
Geovanny Rivadeneyra
José Illanes
Luis A. Salagaje-Muela
Jefferson Gualinga
Tropical forest canopies contain many tree and liana species, and foliar and reproductive characteristics useful for taxonomic identificatio… (voir plus)n are often difficult to see from the forest floor. As such, taxonomic identification often becomes a bottleneck in tropical forest inventories. Here we present a drone-based workflow to automatically acquire large volumes of close-up, ultra-high resolution photos of selected tree crowns (or specific locations over the canopy) to support tropical botanical and ecological studies ( https://youtu.be/80goMEifpc4 ). Our workflow is built around the small, easy-to-use DJI Mavic 3 Enterprise (M3E) drone, which is equipped with a wide-angle and a telephoto camera. On day one, the pilot maps a forest area of up to ∼200 ha with the wide-angle camera to generate a high-resolution digital surface model (DSM) and orthomosaic using structure-from-motion (SfM) photogrammetry. On subsequent days, the pilot acquires close-up photos with the telephoto camera from up to 300 selected canopy trees per day. These close-up photos are acquired from 6 m above the canopy and contain a high level of visual detail that allows botanists to reliably identify many tree and liana species. The photos are geolocated with survey-grade accuracy using RTK GNSS, thus facilitating spatial co-registration with other data sources, including the photogrammetry products. The primary operational challenge of our workflow is the need to maintain RTK corrections with the drone to ensure that close-up photos are acquired exactly at the predefined locations. The maximum operational range we achieved was 3 km, which would allow the pilot to reach any tree within a ∼2800 ha area from the take-off point. Although our workflow was developed to support taxonomic identification of tropical trees and lianas, it could be extended to any other forest or vegetation type to support botanical, phenological, and ecological studies. We provide harpia , an open-source Python library to program these automatic close-up photo missions with the M3E drone ( https://github.com/traitlab/harpia ). We provide harpia , an open-source Python library to program these automatic close-up photo missions ( https://github.com/traitlab/harpia ). Drone imagery and labelled close-up photo data are not yet publicly available because they were acquired with the goal of publishing benchmark machine learning datasets and models for tree and liana species classification and prior publication of the data would jeopardize this future publication.
2026-01-05
bioRxiv (prépublication)
doi.org
Shaped by meaning, weighted by reliability: New insights into multisensory integration
Elizaveta Sycheva
Léa St-Gelais
Karim Jerbi CoCo Lab
Shahab Bakhtiari
Franco Lepore
Vanessa Hadid
2026-01-04
bioRxiv (prépublication)
doi.org
Carriers of LRRK2 pathogenic variants show a milder, anatomically distinct brain signature of Parkinson's disease
Jakub Kopal
Andrew Vo
Qin Tao
Tanya Simuni
Lana M. Chahine
Danilo Bzdok
Alain Dagher
Pathogenic LRRK2 gene variants are a major genetic risk factor for both familial and sporadic Pa… (voir plus)rkinson’s dissease (PD), opening an unattended window into disease mechanisms and potential therapies. Investigating the influence of pathogenic variants in LRRK2 gene on brain structure is a crucial step toward enabling early diagnosis and personalized treatment. Yet, despite its significance, the ways in which LRRK2 genotype affects brain structure remain largely unexplored. Work in this domain is plagued by small sample sizes and differences in cohort composition, which can obscure genuine distinctions among clinical subgroups. In this study, we overcome such important limitations by combining explicit modeling of population background variation and pattern matching. Specifically, we leverage a cohort of 603 participants (including 370 with a PD diagnosis) to examine MRI-detectable cortical atrophy patterns associated with the LRRK2 pathogenic variants in people with PD and carriers without Parkinson’s symptoms. LRRK2 PD patients exhibit milder cortical thinning compared to sporadic PD, with notable preservation in temporal and occipital regions, suggesting a distinct pattern of neurodegeneration. Non-manifesting LRRK2 carriers show no significant cortical atrophy, indicating no structural signs of subclinical PD. We further analyze the relationship between aggregated alpha-synuclein in cerebrospinal fluid and atrophy. We find that those with evidence of aggregated alpha-synuclein experienced pronounced neurodegeneration and increased cortical thinning, possibly defining another aggressive PD subtype. Our findings highlight genetic avenues for distinguishing PD subtypes, which could lead to more targeted treatment approaches and a more complete understanding of Parkinson’s disease progression.
2026-01-02
Communications Medicine (publié)
doi.org
Empowering 2D neural network for 3D medical image segmentation via neighborhood information fusion
Qiankun Li
Xiaolong Huang
Yani Zhang
Bo Fang
Duo Hong
Junxin Chen
2026-01-01
Pattern Recognition (publié)
doi.org
AfriqueLLM: How Data Mixing and Model Architecture Impact Continued Pre-training for African Languages
Hao Yu
Tianyi Xu
Michael A. Hedderich
Wassim Hamidouche
Syed Waqas Zamir
David Ifeoluwa Adelani
2025-12-31
arXiv (publié)
doi.org
arxiv.org
ARM-FM: Automated Reward Machines via Foundation Models for Compositional Reinforcement Learning
Roger Creus Castanyer
Faisal Mohamed
Pablo Samuel Castro
Cyrus Neary
Glen Berseth
Reinforcement learning (RL) algorithms are highly sensitive to reward function specification, which remains a central challenge limiting the… (voir plus)ir broad applicability. We present ARM-FM: Automated Reward Machines via Foundation Models, a framework for automated, compositional reward design in RL that leverages the high-level reasoning capabilities of foundation models (FMs). Reward machines (RMs) - an automata-based formalism for reward specification - are used as the mechanism for RL objective specification, and are automatically constructed via the use of FMs. The structured formalism of RMs yields effective task decompositions, while the use of FMs enables objective specifications in natural language. Concretely, we (i) use FMs to automatically generate RMs from natural language specifications; (ii) associate language embeddings with each RM automata-state to enable generalization across tasks; and (iii) provide empirical evidence of ARM-FM's effectiveness in a diverse suite of challenging environments, including evidence of zero-shot generalization.
2025-12-31
International Conference on Learning Representations (Accept (Poster))
doi.org
openreview.net
Bayesian Deep Learning for Remaining Useful Life Estimation via Stein Variational Gradient Descent
Luca Della Libera
Jacopo Andreoli
Davide Dalle Pezze
Mirco Ravanaelli
Gian Antonio Susto
A crucial task in predictive maintenance is estimating the remaining useful life of physical systems. In the last decade, deep learning has … (voir plus)improved considerably upon traditional model-based and statistical approaches in terms of predictive performance. However, in order to optimally plan maintenance operations, it is also important to quantify the uncertainty inherent to the predictions. This issue can be addressed by turning standard frequentist neural networks into Bayesian neural networks, which are naturally capable of providing confidence intervals around the estimates. Several methods exist for training those models. Researchers have focused mostly on parametric variational inference and sampling-based techniques, which notoriously suffer from limited approximation power and large computational burden, respectively. In this work, we use Stein variational gradient descent, a recently proposed algorithm for approximating intractable distributions that overcomes the drawbacks of the aforementioned techniques. In particular, we show through experimental studies on simulated run-to-failure turbofan engine degradation data that Bayesian deep learning models trained via Stein variational gradient descent consistently outperform with respect to convergence speed and predictive performance both the same models trained via parametric variational inference and their frequentist counterparts trained via backpropagation. Furthermore, we propose a method to enhance performance based on the uncertainty information provided by the Bayesian models. We release the source code at https://github.com/lucadellalib/bdl-rul-svgd.
2025-12-31
IEEE Trans Autom. Sci. Eng. (publié)
doi.org
arxiv.org