Publications

Balancing Stability and Efficiency in Few-Shot Class-Incremental Learning
Abdulmoumen Al-Atrash
Eeham Khan
Ali Ayub
Few-Shot Class-Incremental Learning (FSCIL) requires models to learn new classes from limited samples while retaining prior knowledge under … (see more)strict compute and memory constraints. Existing approaches face a difficult trade-off: fine-tuning is efficient but forgets, replay-based methods mitigate forgetting at high cost, and exemplar-free methods sacrifice adaptability by freezing most of the backbone. We propose Selective Backpropagation (SBP), a deterministic parameter budgeting framework that restricts gradient updates to a pre-allocated subset of parameters, freezing past knowledge and preserving unbiased capacity for future learning without costly mask optimization. SBP achieves state-of-the-art accuracy across standard FSCIL benchmarks while requiring training time close to naive fine-tuning. Crucially, we show that methods relying on frozen backbones fail in cross-domain FSCIL, where distribution shift is unavoidable, while SBP remains highly adaptive and significantly outperforms all competing approaches.
2026-08-24
Work-In-Progress Track @ Conference on Lifelong Learning Agents (poster)
openreview.net
openreview.net
CoPeP: Benchmarking Continual Pretraining for Protein Language Models
Darshan Patil
Pranshu Malviya
Mathieu Reymond
Quentin Fournier
Sarath Chandar
Protein language models (pLMs) have recently gained significant attention for their ability to uncover relationships between sequence, struc… (see more)ture, and function from evolutionary statistics, thereby accelerating therapeutic drug discovery. These models learn from large protein databases that are continuously updated by the biology community and whose dynamic nature motivates the application of continual learning, not only to keep up with the ever-growing data, but also as an opportunity to take advantage of the temporal meta-information that is created during this process. As a result, we introduce the Continual Pretraining of Protein Language Models (CoPeP) benchmark, a novel benchmark for evaluating continual learning approaches on pLMs. Specifically, we curate a sequence of protein datasets derived from the UniProt Knowledgebase spanning a decade and define metrics to assess pLM performance across 31 protein understanding tasks. We evaluate several methods from the continual learning literature, including replay, unlearning, and plasticity-based methods, some of which have never been applied to models and data of this scale. Our findings reveal that incorporating temporal meta-information improves perplexity by up to 7% even when compared to training on data from all tasks jointly. Moreover, even at scale, several continual learning methods outperform naive continual pretraining. The CoPeP benchmark offers an exciting opportunity to study these methods at scale in an impactful real-world application.
2026-08-24
Work-In-Progress Track @ Conference on Lifelong Learning Agents (poster)
doi.org
openreview.net
Extending Ground-Constraint LiDAR-IMU Calibration to Tilted Surfaces in a Continuous-Time Framework
Vassili Korotkine
Pierre Chamoun
Mohammed Ayman Shalaby
James Richard Forbes
This paper presents a novel method that extends targetless LiDAR-IMU calibration for ground vehicles to non- flat environments. Calibration … (see more)typically necessitates full exci- tation of the sensor rig, a requirement that is not fulfilled by ground vehicles in normal operation. To address the degenerate planar motion, state-of-the-art methods propose residuals that assume the colinearity of the gravity and physical surface normal vectors, restricting usage to cases where the ground is assumed flat. This paper proposes ground-plane residuals that do not require this assumption, and are applicable for planar motion on a tilted surface. Results are demonstrated on a dataset collected from a Husky ground vehicle, on the M2DGR dataset, as well as on an offroad vehicle dataset. Repeatability is shown to be improved both in tilted and flat-ground scenarios, with strong improvement demonstrated for the tilted case. The implementation and experiments are open-sourced at https://github.com/vkorotkine/licalib_tilted_ground.
2026-08-24
arXiv (preprint)
doi.org
arxiv.org
Generalization Measures under Controlled Covariate Shift: A Regime-Aware Benchmark
Sora Nakai
Youssef Fadhloun
Kacem Mathlouthi
Kotaro Yoshida
Ganesh Talluri
Ioannis Mitliagkas
Hiroki Naganuma
Predicting generalization from quantities available before target-test evaluation remains a central challenge in deep learning. The systemat… (see more)ic benchmark of Jiang et al. (2020) evaluated many generalization measures, but it focused on independent and identically distributed (IID) settings. We revisit this problem for image classifiers evaluated under controlled corruptions and perturbations. Our study uses CIFAR-10-C/P, where the label space and task remain fixed while the input images are degraded or perturbed. This setting also allows us to revisit the robustness concerns raised by Dziugaite et al. (2020), who showed that the apparent reliability of generalization measures can depend strongly on experimental conditions. Our experiments show that the usefulness of generalization measures is strongly regime-dependent. In our exploratory decision analysis across three CNN-style architectures, sharpness- and input-gradient-based measures are among the leading individual signals, whereas family results are close and architecture dependent. Optimization-based measures, Information Criteria, and Sharpness-based measures provide additional regime-dependent signals in correlation or local-reliability analyses. Together, these findings suggest that model selection should not rely only on measures favored by IID evaluation. Instead, within the evaluated CIFAR-10-C/P setting and architectures, generalization measures should be treated as regime-dependent ranking signals whose utility must be evaluated for the intended corruption or perturbation setting.
2026-08-24
Transactions on Machine Learning Research (accepted)
doi.org
openreview.net
Loss Smoothing for Stable Adaptation Under Distribution Shift
Darshan Patil
Ekaterina Lobacheva
Razvan Pascanu
Sarath Chandar
In settings such as fine-tuning and reinforcement learning, neural networks are often adapted under distribution shift. Standard adaptation … (see more)methods typically optimize the target objective directly, inducing an abrupt change from the source training objective. This abrupt transition can distort learned representations, including features that may still be useful for the new task. We investigate whether a more gradual transition can improve adaptation. We propose loss smoothing, a simple approach that interpolates between the source and target training objectives at the start of adaptation. This smooth transition helps to preserve useful features from the source distribution while still enabling the model to specialize to the target distribution. Across controlled supervised shifts, pretrained vision adaptation, offline-to-online and online reinforcement learning, and language model fine-tuning, we find that loss smoothing consistently improves performance, suggesting that smoother objective transitions are a broadly useful tool for model adaptation.
2026-08-24
Work-In-Progress Track @ Conference on Lifelong Learning Agents (poster)
doi.org
openreview.net
Network-level functional connectivity is associated with longitudinal tau accumulation and amyloid-dependent cognitive decline in preclinical Alzheimer’s disease
Mohammadali Javanray
Jonathan Gallego-Rudolf
Yara Yakoub
Ting Qiu
Frederic St‐Onge
Jordana Remz
Jean-Paul Soucy
Bratislav Misic
Lune Bellec
Jacob Vogel
Sylvia Villeneuve
2026-08-24
Research Square (published)
doi.org
Overcoming Rank Collpase in Feedback Alignment
Gauthier Boeshertz
Claudia Clopath
Razvan Pascanu
Backpropagation (BP) is widely viewed as biologically implausible, in part because it requires feedback weights to be the transpose of forwa… (see more)rd weights for error propagation. Interestingly, when training a network with fixed random feedback weights to circumvent this issue, learning aligns the forward weights with the feedback weights, leading the backpropagated error signal to become an approximation of the standard gradient used by BP. This process, called Feedback Alignment (FA), occurs in MLPs and very shallow CNNs but does not scale well to deeper architectures. In this work, we first investigated differences between BP and FA models, trained on CIFAR10, specifically focusing on the effective rank of the signal. We found that the FA error has a considerably lower rank and hence is constrained to a lower-dimensional subspace compared to BP, limiting exploration of the parameter space. Motivated by this observation, we evaluated two mechanisms for increasing the effective dimensionality of FA: Muon, an optimiser that orthogonalises weight updates; and hidden activity normalisation, which promotes activation orthogonality. Across larger architectures and benchmarks, we find that these methods consistently improve over FA baselines, for example, on CIFAR100 with a Resnet-18, accuracy increases by 9 percentage points. Our results identify low-dimensional gradient dynamics as a key obstacle to scaling FA and suggest that inducing higher-dimensional update geometry is a promising route toward scaling alternatives to backpropagation.
2026-08-24
Work-In-Progress Track @ Conference on Lifelong Learning Agents (poster)
openreview.net
openreview.net
Mapping Alzheimer’s neuropathology signatures to the whole brain transcriptome using machine learning data-fusion
Anwesha Bhattacharya
Chloé Savignac
L. M. Hodgson
Jack Stanley
Guy Wolf
Smita Krishnaswamy
David A Bennett
Elisabeth B. Binder
Danilo Bzdok
Abstract In Alzheimer’s disease (AD), misfolded proteins emerge across the entire brain in structured, yet not rigid, spatiotemporal patte… (see more)rns. Yet, a systematic bias of single-cell genomics toward sampling mostly cortical tissue limits our understanding of the whole-brain transcriptomic vulnerability to AD. Here, we develop a machine learning method to extrapolate local AD neuropathology signatures to the whole brain. By analyzing gene expression profiles of over two million cortical cells from 427 humans spanning the AD-pathology spectrum, we derive transcriptomic estimators of AD neuropathology. After extensive validations on datasets with known ground truth, we apply this framework to three million cells from 108 brain regions in the Siletti whole human brain atlas and derive an anticipated brain map of transcriptomic signatures indexing AD neuropathology. This interrogation of regions spanning the cortical, subcortical, and brainstem structures uncovers transcriptomic signatures associated with hyperphosphorylated tau in the medulla oblongata, dorsal raphe nucleus, and the tuberal and mammillary regions of the hypothalamus. At the cellular level, assessments of these signatures across 31 cell populations identify VGLUT1/2 expressing neurons, astrocytes, and microglia as key neuropathology-resembling populations. Within the hippocampus, pathology signatures surface in the rostral cornu ammonis (CA) subfields, particularly in the CA1 pyramidal neurons and dentate granule cells. β-amyloid-like signatures localize to the neocortex with laminar selectivity — most prominently in upper layer somatostatin+ intratelencephalic neurons (L2-L3), but also in deep layer intratelencephalic and corticothalamic neurons (L5-L6). Neocortical astrocytes and microglia exhibiting disease associated signatures similarly demonstrate a unique laminar preference. Together, this study provides the first whole human brain map of AD pathology-associated transcriptomic signals, and exposes cell type, region, and cortex layer specific vulnerabilities.
2026-08-23
bioRxiv (accepted)
doi.org
Spatiotemporal Distillation via Recurrent Bottlenecks for Aortic Tracking
Dexter Wen Jie Teo
Nairouz Shehata
Hervé Lombaert
Cardiac cine-MRI serves as a direct visual indicator of cardiovascular hemodynamics by capturing the continuous wall motion of the aorta. Qu… (see more)antifying these dynamic structural changes across the cardiac cycle is essential for measuring aortic distensibility, a primary marker of arterial stiffness. However, standard 2D segmentation networks focus on each frame independently. Consequently, when rapid systolic flow temporarily obscures the aorta's boundaries, this lack of continuous context results in frame-to-frame tracking dropouts and boundary inconsistencies. Spatiotemporal (
2026-08-23
arXiv (preprint)
doi.org
arxiv.org
Configurational Approaches to Optimal Distinctiveness: Exploring Signal Sets and Originality for Entrepreneurship Legitimacy
Maxime Mellard
Romain Rampa
Guy Parmentier
Florian Carichon
For entrepreneurs, the legitimisation process in the early stages is crucial for acquiring essential resources and ensuring the viability an… (see more)d growth of their new ventures. Drawing on signalling theory and optimal distinctiveness literature, the article investigates the configurations of signals that entrepreneurs can send in the early stages of their original entrepreneurial project to attain various thresholds of legitimacy. The study is based on a Fuzzy-Set Qualitative Comparative Analysis (fsQCA) analysis of 145 crowdfunding projects. The findings reveal that original projects require more intricate signal configurations than non-original projects to achieve high levels of legitimacy. The present research advances the ongoing discourse by unveiling the dynamic interplay between distinctiveness, legitimacy, and signal configurations.
2026-08-22
Journal of Entrepreneurship (published)
doi.org
Evaluation Awareness in Language Models: Representation, Verbalization, and Control
Farzaneh Heidari
Amin Memarian
Guillaume Rabusseau
Both capability and safety benchmarks rest upon the assumption that the behavior of language models undergoing a test is informative about t… (see more)heir behavior in deployment. This assumption can fail, should models infer that they are being evaluated and condition their response on such context. This hypothesis, termed ``evaluation awareness'', has been observed in frontier and open-weight language models alike. We provide a systematic study of this phenomenon, by probing for it across six language models (from four families and three sizes) and three metrics. More precisely, we examine whether (i) being under evaluation is linearly represented within the models'activations space, (ii) it is verbalized in their output tokens (as scored by an LLM-as-judge), and (iii) steering causally affects their behavior. For the open-checkpoint Olmo models, we further test these measures at every training stage. In doing so, we report that evaluation awareness is linearly decodable from the residual streams of every model (best AUROC
2026-08-21
arXiv (preprint)
doi.org
arxiv.org
BiDRA: Bayesian Inference as a Robust Alternative to Non-Linear Regres-sion for Dose-Response Efficiency Metrics Assessment
Caroline Labelle
Petr Smirnov
Maud David
Mario Callejo
Benjamin Haibe-Kains
Anne Marinier
Sébastien Lemieux
Abstract Motivation Dose-response metrics such as the half-maximal inhibitory concentration, high-dose response, and slope are central to dr… (see more)ug discovery, yet standard Levenberg-Marquardt fits often produce biased or unsupported values for incomplete curves and lack uncertainty measures. In practice, this forces experimenters to visually inspect each curve fit to judge its reliability, a tedious, subjective, and non-scalable process. Results Across 421,405 public dose-response experiments from three large pharmacogenomic datasets, shared-concentration viability responses were highly consistent across biological replicates, making it reasonable to expect derived efficiency metrics to show comparable replicate behavior when supported by the data. However, Levenberg-Marquardt fits to incomplete curves often produced unsupported metric values, including inflated rates of observable potency estimates (>70% versus ∼39% complete curves), which could falsely suggest metric-level disagreement between similar replicate responses. BiDRA represents efficacy, potency, and slope as posterior distributions, allowing uncertainty to remain large when the data do not support precise metric inference. This avoids treating uncertainty-dominated responses as conflicting point estimates, exposes unsupported Levenberg-Marquardt estimates, and supports uncertainty-aware compound ranking. We illustrate its utility in a structure–activity relationship screen, demonstrating how posteriors enable robust, criteria-based selection in drug discovery settings. Availability and implementation BiDRA is implemented in Julia and available at https://github.com/lemieux-lab/bidra_robustness.
2026-08-20
Bioinformatics Advances (published)
doi.org