Publications

Debiased One-Shot NAS Via Density-Aware Sampling
Mehraveh Javan Roshtkhari
Matthew Toews
One-shot Neural Architecture Search (NAS) is based on training a supernet, a single model from which many different architectures with share… (see more)d weights are sampled and updated during training. Training the shared weights of the supernet is much more computationally efficient than training each architecture independently. However, during the supernet training, architectures with similar gradients (dense regions in the gradient space) would cooperate with each other increasing their training, while architectures in sparse(low-density) regions would not receive as much training benefit from others. This does not allow all architectures to be trained with the same amount of effective updates, producing a training bias that favors architectures in denser regions. As a consequence, the correlation between the supernet estimations and the actual performance of models independently trained is reduced. This negatively affects the supernet's ability to select good architectures once trained. In this paper, we propose two computationally feasible ways for different computational budgets and search spaces to approximate the architecture densities and implement a density-aware debiasing mechanism for supernet training. We propose a fine-grained density calculation for numerable smaller search spaces and an online density approximation based on density prototypes from a clustering algorithm for larger spaces. We validate our method on CIFAR10, CIFAR100 and ImageNet datasets using various search strategies and show consistently improved results compared to several single-path one-shot supernet training methods.
FACT-GS: Frequency-Aligned Complexity-Aware Texture Reparameterization for 2D Gaussian Splatting
Tianhao Xie
Xinxin Zuo
Tiberiu Popa
Realistic scene appearance modeling has advanced rapidly with Gaussian Splatting, which enables real-time, high-quality rendering. Recent ad… (see more)vances introduced per-primitive textures that incorporate spatial color variations within each Gaussian, improving their expressiveness. However, texture-based Gaussians parameterize appearance with a uniform per-Gaussian sampling grid, allocating equal sampling density regardless of local visual complexity, which leads to inefficient texture space utilization. We introduce FACT-GS, a Frequency-Aligned Complexity-Aware Texture Gaussian Splatting framework that allocates texture sampling density according to local visual frequency. Grounded in adaptive sampling theory, FACT-GS reformulates texture parameterization as a differentiable sampling-density allocation problem, replacing the uniform textures with a learnable frequency-aware allocation strategy implemented via a deformation field whose Jacobian modulates local sampling density. Built on 2D Gaussian Splatting, FACT-GS performs non-uniform sampling on fixed-resolution texture grids, preserving real-time performance while recovering sharper high-frequency details under the same parameter budget.
Genetic landscape of an in vivo protein interactome
Savandara Besse
Tatsuya Sakaguchi
Louis Gauthier
Zahra Sahaf
Rini Ravindran
Lidice Gonzalez
Xavier Castellanos-Girouard
Cassandre Clermont
Nazli Koçatug
Olivier Péloquin
Chloé Matta
Stephen W. Michnick
Adrian W.R. Serohijos
Learning Through Creation: A Hash-Free Framework for On-the-Fly Category Discovery
Bohan Zhang
Weidong Tang
Zhixiang Chi
Yi Jin
Zhenbo Li
Yanan Wu
On-the-Fly Category Discovery (OCD) aims to recognize known classes while simultaneously discovering emerging novel categories during infere… (see more)nce, using supervision only from known classes during offline training. Existing approaches rely either on fixed label supervision or on diffusion-based augmentations to enhance the backbone, yet none of them explicitly train the model to perform the discovery task required at test time. It is fundamentally unreasonable to expect a model optimized on limited labeled data to carry out a qualitatively different discovery objective during inference. This mismatch creates a clear optimization misalignment between the offline learning stage and the online discovery stage. In addition, prior methods often depend on hash-based encodings or severe feature compression, which further limits representational capacity. To address these issues, we propose Learning through Creation (LTC), a fully feature-based and hash-free framework that injects novel-category awareness directly into offline learning. At its core is a lightweight, online pseudo-unknown generator driven by kernel-energy minimization and entropy maximization (MKEE). Unlike previous methods that generate synthetic samples once before training, our generator evolves jointly with the model's dynamics and synthesizes pseudo-novel instances on-the-fly at negligible cost. These samples are incorporated through a dual max-margin objective with adaptive thresholding, strengthening the model's ability to delineate and detect unknown regions through explicit creation. Extensive experiments across seven benchmarks show that LTC consistently outperforms prior work, achieving improvements ranging from 1.5% to 13.1% in all-class accuracy.
LLJ Cards: Best practices for the Use of LLMs as Judges
Melina Medjdoub
Jackie Chi Kit Cheung
In recent years, large language models (LLMs) have emerged as a popular alternative for evaluation. Often referred to as LLMs as judges (LLJ… (see more)s), these systems have been widely adopted by researchers and practitioners across a broad range of measurement tasks, driven by their strong performance, scalability, and cost-effectiveness relative to human judgment. However, a growing body of work has shown that the use of LLJs raise concerns about their validity and reliability as evaluators. Existing efforts to address these challenges have largely focused on developing bias-mitigation techniques and refining prompting strategies. While these approaches represent an important step forward, they primarily offer technical fixes and leave a more fundamental challenge unaddressed: the lack of standardized, transparent, and reproducible evaluation practices. In this paper, we introduce LLJ Cards, a framework that synthesizes best practices from measurement theory, natural language generation, and machine learning literature into practical guidelines for LLJ-based evaluations. While LLJs offer a promising path toward scalable evaluation, their effective use requires grounding in rigorous evaluation principles to ensure validity, reliability, and reproducibility. LLJ Cards addresses this need by providing a structured framework for applying these principles in the design and reporting of automated evaluations.
Paired neuronal-haemodynamic neurovascular coupling measures and cognitive performance in vascular pathology: protocol for a systematic review and meta-analysis
V. D. Abramova
Marta Estrada
Veronica Egovtseva
Ksenya Pronyaeva
Rustam Talybov
Taleb M. Almansoori
Mohammad I. K. Hamad
Milos Ljubisavljevic
Yauhen Statsenko
Abstract Background Neurovascular coupling (NVC) links neuronal activity to haemodynamic responses. Altered NVC may contribute to cognitive … (see more)impairment in vascular pathology, but evidence is limited by inconsistent definitions and by the conflation of paired neuronal–haemodynamic NVC measures with haemodynamic-only, model-derived, or imaging-derived proxy measures. Aim and objectives The primary aim is to evaluate the association between paired neuronal–haemodynamic NVC measures and cognitive performance in middle-aged and older adults with vascular pathology. Secondarily, we will compare paired neuronal–haemodynamic NVC measures between patients with vascular pathology and age-matched controls without vascular pathology, determine whether predefined pathology strata show distinct impairment profiles, and examine whether the NVC–cognition association varies across cognitive domains. As an exploratory objective, we will synthesise evidence on the prognostic utility of paired neuronal–haemodynamic NVC measures in predicting future cognitive decline. Collectively, these objectives address critical gaps in a fragmented literature characterised by heterogeneous methodologies, populations, and assessment techniques, as well as methodological limitations including small samples, cross-sectional designs, and inadequate confounder control. This review will systematically synthesise existing findings, identify remaining knowledge gaps, and outline priorities for future research. Methods and analysis This protocol follows PRISMA-P guidance and is registered with PROSPERO (CRD420261351976). We searched five electronic databases from inception to 28 July 2026 for peer-reviewed English-language studies. Eligible studies include middle-aged and older adults with vascular pathology and/or participants from relevant comparator groups and report paired neuronal–haemodynamic NVC measures based on independently acquired neuronal and haemodynamic or vascular signals. The primary synthesis will evaluate associations between paired neuronal–haemodynamic NVC measures and cognitive performance, with meta-analysis conducted only for synthesis strata that meet prespecified feasibility criteria.
A principled approach for energy-efficient training via phase-aware GPU frequency tuning
Miguel Braga
Júlio Pinto
Olivier Michaud
Cláudia Brito
Ricardo Macedo
Modern AI model training imposes unprecedented computational demands, making it a key contributor to datacenter energy consumption. Yet a si… (see more)gnificant fraction of the energy consumed during training does not translate to useful computation due to bottlenecks throughout the training pipeline. We present PAFT, a phase-aware, dynamically adaptable GPU frequency tuning system that reduces energy consumption of training workloads with minimal performance overhead. The key insight behind PAFT is that bottlenecks represent an energy optimization opportunity, rather than purely a performance problem: when GPUs are bound to stall, PAFT opportunistically reduces their clock frequencies to match the pace of bottlenecked devices, saving energy without impacting execution time. PAFT achieves this by continuously monitoring pipeline behavior and applying fine-grained frequency adjustments, adapting to workload and system changes. Experiments conducted on twelve widely used models show that PAFT consistently outperforms all baselines, achieving energy savings of up to 46% with an average overhead of 4%.
Range-Aided SLAM Initialization Exploiting Accurate Heading Information
This paper presents a novel initialization method for range-aided simultaneous localization and mapping (RA-SLAM). The general SLAM problem … (see more)has a well-known separable structure where landmark and robot positions can be solved for in a linear fashion given known robot headings. This paper considers the case where highly accurate heading information is available, which is typical of autonomous underwater vehicle (AUV) navigation, to solve for the range transponder and robot positions. The proposed approach consists of two steps. First, using a generalized trust region subproblem (GTRS), the positions of the transponders relative to the AUV are solved for given the nonlinear range measurements. Second, the relative transponder positions and the known heading of the AUV are used to estimate the transponder and AUV positions by solving a linear least-squares problem. These transponder and AUV position estimates, combined with the highly accurate heading information, provide a reliable initialization method for the general nonlinear RA-SLAM problem. The effectiveness of the proposed approach is tested on a real-world AUV dataset where long baseline (LBL) range measurements are provided in concert with highly accurate heading information provided by an inertial navigation system (INS).
Revisiting Forward-Looking Sonar Image Registration
Forward-looking sonar is a common underwater sensing modality. Sonar-image registration is a crucial process in many sonar-aided navigation … (see more)tasks. Since the last comparison made between forward-looking sonar image registration methods, different methods have been proposed in the literature, as well as modern point-feature extractors that can be used for registration. An updated comparison is performed both on simulated and experimental sonar-image registration tasks. It is shown that Radon-transformation-based registration outperforms all other registration methods in terms of accuracy and robustness, especially when there is a large change in position between sonar images.
Time-Incremental Continued Pretraining of LLMs: Knowledge Updates Without Catastrophic Forgetting
Salman Hussain Ali
Çağatay Yıldız
Large language models (LLMs) drift out of date the moment their pretraining ends, yet retraining from scratch is prohibitively expensive. Co… (see more)ntinued pretraining (CPT) is the natural remedy, but it is typically evaluated through a continual learning lens that assumes disjoint data streams. This is a poor fit for time-incremental updates on web-scale crawls, where successive snapshots share substantial URL overlap by design. We study time-incremental CPT in this realistic regime: continued pretraining on FineWeb-Edu dumps drawn strictly from after each model's knowledge cutoff, evaluated across six open-weight models spanning three families (OLMo2, Llama-3.1/3.2, Gemma-3-1B) and four parameter scales (1B-3B-7B-8B). We organize our findings around four practical questions. (i) Is knowledge acquired? Yes, but heterogeneously, and without catastrophic forgetting: five of six models also improve on pre-cutoff factual recall, and the gains track pretraining saturation (driven primarily by token budget per parameter). (ii) What does it cost? Almost nothing: the macro-average across a thirteen-task suite stays within 0.01 of the base for every model. (iii) What is the recipe? Data quality dominates quantity (a curated 6B-token slice matches a broader 40B one); the optima for knowledge acquisition and general capability are separated by roughly an order of magnitude in learning rate; and LoRA at sufficient rank matches full CPT. (iv) Does it survive deployment? CPT gains transfer through SFT, while DPO's effect is family-dependent. Together, these results paint a more optimistic picture of time-incremental CPT than the prior continual learning literature suggests.
On the Control of Mobile Ad-Hoc Agent Deployments in Partially Observed Space
Louis-Roy Langevin
Shuo Wen
Ndiamé Ndiaye
Antonio Loría
We study the online deployment of mobile ad hoc networks in unknown orthogonal environments, formalized as the Partially Observable Cooperat… (see more)ive Guard Art Gallery Problem. We give a full proof that CADENCE algorithms achieve full coverage while maintaining a connected visibility graph using at most
BrainWideBench: Benchmarking large-scale pretraining and across-animal transfer in multi-region neural recordings
Alexandre Andre
Shivashriganesh P. Mahato
Vinam Arora
Keshav Balaji
Nanda H. Krishna
Jingyun Xiao
Yizi Zhang
Wenrui Ma
Han Yu
International Brain Laboratory
Daniel Birman
Niccolò Bonacchi
Gaelle A. Chapuis
Joana A. Catarino
Felicia Davatolhagh
Mayo Faulkner
Laura Freitas-Silva
Fei Hu … (see 22 more)
Julia M. Huntenburg
Anup Khanal
Inês Laranjeira
Petrina Lau
Guido T. Meijer
Nathaniel J. Miska
Jean-Paul Noel
Alejandro Pan-Vazquez
Georg Raiser
Cyrille Rossant
Karolina Z. Socha
Anne E. Urai
Miles J. Wells
Steven J. West
Olivier Winter
Cole Hurwitz
Mehdi Azabou
Matthew R. Whiteway
Liam Paninski
Eva L. Dyer
Advances in large-scale neural recording have made it possible to collect data across many animals and distributed brain regions, raising th… (see more)e question of whether this scale can be exploited to learn general-purpose neural representations transferable across diverse downstream tasks. Yet, progress toward this goal has been limited by fragmented evaluation protocols and a narrow focus on individual task domains. Here, we present BrainWideBench, a benchmark for evaluating across-animal transfer on multi-region neural recordings, built on the International Brain Laboratory Brainwide Map dataset of neural and behavioral recordings spanning 276 brain regions from 139 mice performing a sensory-guided decision-making task. The benchmark is organized around three complementary task suites that evaluate whether learned representations support downstream decoding of behavior, can predict masked or future neural activity, and can recover biologically meaningful anatomical organization. With this benchmark, we systematically evaluate pretraining methods across transfer settings, including finetuning on downstream objectives and zero-shot generalization to unseen animals. Our results confirm pretraining improves performance over matched single-session baselines, but we show current methods exhibit heterogeneity in transfer capabilities: gains depend strongly on the alignment between pretraining objectives and downstream tasks. No single approach performs uniformly well across all three suites, and most methods are designed to only address a subset of them. Together, these findings suggest that learning representations that jointly generalize across behavior, dynamics, and anatomy remains an open challenge. By providing a unified and reproducible evaluation suite, BrainWideBench establishes a framework for measuring progress toward general-purpose models of the mouse brain.