Publications

Neurobagel: building an international network for distributed data discovery
Michelle Wang
Jean-Baptiste Poline
Yaroslav O. Halchenko
Jan G. Bjaalie
Katie M. Lavigne
Jeffrey Grethe
Max A. Laansma
Barbara Strasser-Kirchweger
Emile d’Angremont
David N. Kennedy
Neda Jahanshad
Sean N. Hatton
Nikhil Bhagwat
Tristan Glatard
Brent McPherson
Satrajit Ghosh
Gabriel Devenyi
Stéphane Lehéricy
Vincent Taschereau‐Dumouchel
Florian Hutzler … (see 18 more)
Sebastian Urchs
Michael Hanke
Christopher J. Markiewicz
Russell A. Poldrack
Francis Jeanson
Eva van Heese
David Keator
Camille Maumet
M. Mallar Chakravarty
Franco Pestilli
Julia-Katharina Pfarr
Erin W Dickie
Alyssa Dai
Arman Jahanpour
Mathieu Dugré
Lyuba Zehl
Ysbrand van der Werf
Paul Thompson
International data privacy regulations impede the pooling of research data for collaborative analysis. We introduce Neurobagel, a federated … (see more)network enabling cohort discovery across locally governed, access-controlled datasets. Through intuitive graphical tools and a decentralized query infrastructure, Neurobagel facilitates harmonization, control, and discovery of data according to local regulations. Today, Neurobagel is deployed by consortia and data platforms in Europe, North America, Asia, and Australia, supporting diverse and evolving regulatory frameworks.
2026-05-07
MetArXiv (preprint)
doi.org
RNAGenScape: Property-Guided, Optimized Generation of mRNA Sequences with Manifold Langevin Dynamics.
Danqi Liao
Chen Liu
Xingzhi Sun
Dié Tang
Haochen Wang
Scott Youlten
Srikar Krishna Gopinath
Haejeong Lee
Ethan C. Strayer
Antonio J. Giraldez
Smita Krishnaswamy
Generating property-optimized mRNA sequences is central to applications such as vaccine design and protein replacement therapy, but remains … (see more)challenging due to limited data, complex sequence-function relationships, and the narrow space of biologically viable sequences. Generative methods that drift away from the data manifold can yield sequences that fail to fold, translate poorly, or are otherwise nonfunctional. We present RNAGenScape, a property-guided manifold Langevin dynamics framework for mRNA sequence generation that operates directly on a learned manifold of real data. By performing iterative local optimization constrained to this manifold, RNAGenScape preserves biological viability, accesses reliable guidance, and avoids excursions into nonfunctional regions of the ambient sequence space. The framework integrates three components: (1) an autoencoder jointly trained with a property predictor to learn a property-organized latent manifold, (2) a denoising autoencoder that projects updates back onto the manifold, and (3) a property-guided Langevin dynamics procedure that performs optimization along the manifold. Across three real-world mRNA datasets spanning two orders of magnitude in size, RNAGenScape increases median property gain by up to 148% and success rate by up to 30% while ensuring biological viability of generated sequences, and achieves competitive inference efficiency relative to existing generative approaches.
2026-05-07
arXiv (published)
pubmed.ncbi.nlm.nih.gov
Rotation-Preserving Supervised Fine-Tuning
Hangzhan Jin
Tianwei Ni
Lu Li
Pierre-Luc Bacon
Mohammad Hamdaqa
Doina Precup
Supervised fine-tuning (SFT) improves in-domain performance but can degrade out-of-domain (OOD) generalization. Prior work suggests that thi… (see more)s degradation is related to changes in dominant singular subspaces of pretrained weight matrices. However, directly identifying loss-sensitive directions with Hessian or Fisher information is computationally expensive at LLM scale. In this work, we propose preserving projected rotations in pretrained singular subspaces as an efficient proxy for Fisher-sensitive directions, which we call Rotation-Preserving Supervised Fine-Tuning (RPSFT). RPSFT penalizes changes in the projected top-
2026-05-07
arXiv (preprint)
doi.org
openreview.net
Simply the best – A systematic evaluation approach for third-party libraries based on mobile app quality attributes
Rubén Saborido
Rémy Raes
Rodrigo Morales
Romain Rouvoy
Foutse Khomh
Yann-Gaël Guéhéneuc
Abstract Mobile device applications (apps) are complex because they rely on integrating multiple third-party libraries (TPLs). Yet, TPLs eas… (see more)e app development by offering implementations of specific functionality. For example, app developers often use advertising libraries to generate revenue, integrate social networking libraries to simplify login, or include crash reporting libraries to monitor/report crashes in their apps. However, there are multiple TPLs with similar functionalities from which to choose, and developers often cannot foresee all the consequences of using these libraries in their apps. The sizes of apps grow with the addition and usage of TPLs, and so does the number of required permissions and resource consumption. Thus, TPLs may degrade the quality of apps and developers need help measuring and comparing them. We propose EQuAT, an approach for Evaluating Quality Attributes of TPLs that eases the comparison of TPLs. EQuAT takes as input minimal apps that integrate TPLs and playable scenarios to simulate user interaction while exercising a particular functionality of the included TPL. By collecting quality metrics and comparing them using plots, we provide app developers with a systematic approach to rank TPLs based on their preferences. We show how EQuAT helps developers make informed decisions about which libraries to integrate into their apps by validating them against nine TPLs across three categories.
2026-05-07
Empirical Software Engineering (published)
doi.org
TCMIIES: A Browser-Based LLM-Powered Intelligent Information Extraction System for Academic Literature
Hanqing Zhao
The exponential growth of academic publications has created an urgent need for automated tools capable of extracting structured knowledge fr… (see more)om unstructured scientific texts. While large language models (LLMs) have demonstrated remarkable capabilities in natural language understanding and information extraction, existing solutions often require specialized infrastructure, programming expertise, or fine-tuned domain-specific models that create barriers for researchers in specialized fields. This paper presents TCMIIES, a browser-based, zero-installation platform that leverages commercial LLM APIs to perform structured information extraction from academic literature. The system employs a novel schema-guided prompting framework with automatic system prompt generation, enabling researchers to define custom extraction schemas through an intuitive graphical interface without any programming. TCMIIES features a pure front-end architecture that ensures data privacy by processing all information locally in the browser, supports five major LLM providers, implements concurrent batch processing with automatic retry mechanisms, and provides intelligent field mapping for Chinese academic databases including CNKI and Wanfang. We demonstrate the system's effectiveness through comprehensive evaluation across multiple extraction scenarios in Traditional Chinese Medicine research, achieving structured output compliance rates exceeding 94\% and information extraction accuracy comparable to domain-expert annotation. The system represents a practical, accessible solution that bridges the gap between advanced LLM capabilities and domain-specific academic information extraction needs, particularly for researchers in specialized fields who require flexible, privacy-preserving, and cost-effective extraction tools.
2026-05-07
arXiv (preprint)
doi.org
arxiv.org
Agentick: A Unified Benchmark for General Sequential Decision-Making Agents
Roger Creus Castanyer
Pablo Samuel Castro
Glen Berseth
AI agent research spans a wide spectrum: from RL agents that learn from scratch to foundation model agents that leverage pre-trained knowled… (see more)ge, yet no unified benchmark enables fair comparison across these approaches. We present Agentick, a benchmark for sequential decision-making agents designed to evaluate RL, LLM, VLM, hybrid, and human agents on common ground and to power research on the fundamental challenges of sequential decision-making. Agentick provides 37 procedurally generated tasks across six capability categories, four difficulty levels, and five observation modalities, all exposed through a single Gymnasium-compatible interface. The benchmark ships with a Coding API, oracle reference policies for all tasks, pre-built SFT datasets, a composable agent harness, and a live leaderboard. An evaluation spanning 27 configurations and over 90,000 episodes reveals that no single approach dominates: GPT-5 mini leads overall at 0.309 oracle-normalized score while PPO dominates planning and multi-agent tasks; the reasoning harness multiplies LLM performance by 3-10x; and ASCII observations consistently outperform natural language. These findings highlight the substantial room for improvement that remains across all agent paradigms. Agentick's capability-decomposed, multi-modal design provides the empirical infrastructure needed to drive progress toward general autonomous agents, both as an evaluation framework and as a training ground for RL post-training of foundation models in truly sequential environments.
2026-05-06
arXiv (preprint)
doi.org
arxiv.org
Beyond the total NIHSS score: association between impaired level of consciousness and early neurological deterioration in mild large vessel occlusion stroke
Qiangze Ji
Liangliang Sun
Zenghui Liu
Hanqing Zhao
Jing Yu
Ying Zhang
Kaiyue Duan
Lili Guo
Qiuyi Zhang
2026-05-06
Neuroradiology (published)
doi.org
Diversity Curves for Graph Representation Learning
Katharina Limbeck
Nadja Häusermann
Martin Carrasco
Guy Wolf
Bastian Rieck
Graph-level representations are crucial tools for characterising structural differences between graphs. However, comparing graphs with diffe… (see more)rent cardinalities, even when sampled from the same underlying distribution, remains challenging. Unsupervised tasks in particular require interpretable, scalable, and reliable size-aware graph representations. Our work addresses these issues by tracking the structural diversity of a graph across coarsening levels. The resulting graph embeddings, which we denote diversity curves, are interpretable by construction, efficient, and directly comparable across coarsening hierarchies. Specifically, we track the spread of graphs, a novel isometry invariant that is inherently well-suited for encoding the metric diversity and geometry of graphs. We utilise edge contraction coarsening and prove that this improves expressivity, thus leading to more powerful graph-level representations than structural descriptors alone. Demonstrating their utility over a range of baseline methods in practice, we use diversity curves to (i) cluster and visualise simulated graphs across varying sizes, (ii) distinguish the geometry of single-cell graphs, (iii) compare the structure of molecular graph datasets, and (iv) characterise geometric shapes.
2026-05-06
arXiv (preprint)
doi.org
arxiv.org
Kurtosis-Guided Denoising Score Matching for Tabular Anomaly Detection
Victor Livernoche
Jie Zan
Reihaneh Rabbany
Denoising score matching (DSM) provides a way to learn data distributions by training a neural network to recover the score function, define… (see more)d as the gradient of the log density, from noise-corrupted samples. Once trained, the score magnitude at a test point reflects how consistent that point is with the learned distribution, making it a natural anomaly signal. The key practical challenge is selecting the perturbation scale: too little noise yields unstable score estimates in sparse regions, while too much erases local structure and weakens anomaly sensitivity. This is compounded by the difficulty of hyperparameter tuning when anomalies are unknown and no validation set is available. We introduce kurtosis-based noise scaling (K-DSM), a per-feature scheme that sets noise levels from the shape of each marginal distribution, improving coverage of low-density regions and precision in high-density regions without extra model complexity. Contrary to prior claims that multi-scale or noise-conditioned training is necessary, we find that a carefully trained single-scale model is already a strong anomaly detector. On standard tabular anomaly detection benchmarks, K-DSM achieves state-of-the-art performance in the semi-supervised setting. When combined with a lightweight EMA-teacher filtering rule that removes low-density training points before each gradient step, it also achieves strong performance in the fully unsupervised (contaminated) setting, suggesting that simple, data-adaptive noise scaling enables robust anomaly detection while reducing reliance on hyperparameter tuning.
2026-05-06
arXiv (preprint)
doi.org
arxiv.org
Orth-Dion: Eliminating Geometric Mismatch in Distributed Low-Rank Spectral Optimization
Tatsuhiro Nakamori
Laura Gomezjurado Gonzalez
Ganesh Talluri
Ansh Tiwari
Hideyuki Kawashima
Ioannis Mitliagkas
Guillaume Rabusseau
Hiroki Naganuma
Low-rank gradient compression reduces communication in distributed training by representing updates with rank-…
2026-05-06
arXiv (preprint)
doi.org
arxiv.org
Revisiting Adam for Streaming Reinforcement Learning
Florin Gogianu
Adrian Catalin Lutu
Razvan Pascanu
Learning from a sequence of interactions, as soon as observations are perceived and acted upon, without explicitly storing them, holds the p… (see more)romise of simpler, more efficient and adaptive algorithms. For over a decade, however, deep reinforcement learning walked the contrary path, augmenting agents with replay buffers or parallel sampling routines, in an effort to tame learning instability. Recently, this topic has been revisited by Elsayed et al. (2024), focusing on update computation through eligibility traces and modifications to the optimisation routine, resulting in the StreamQ algorithm. In this work we take a step back, investigating the efficacy of established updates, such as those implemented by DQN and C51 within this online setting. Not only do we find that they perform well, but through analysing how the optimisation algorithm generally, and Adam in particular, interacts with these updates, we contend that two properties are essential for robust performance: i) the derivative of the objective is to be bounded and ii) weight updates are variance-adjusted. Rigorous and exhaustive experimentation demonstrates that C51, which exhibits both characteristics, is competitive with StreamQ across a subset of 55 Atari games. Using these insights, we derive a variance-adjusted algorithm based on eligibility traces, termed Adaptive Q
2026-05-06
arXiv (preprint)
doi.org
arxiv.org
Spectral Lens: Activation and Gradient Spectra as Diagnostics of LLM Optimization
Andy Zeyi Liu
Elliot Paquette
John Sous
Training loss and throughput can hide distinct internal representation in language-model training. To examine these hidden mechanics, we use… (see more) spectral measurements as practical and operational diagnostics. Using a controlled family of decoder-only models adapted from the modded NanoGPT codebase, we introduce an empirical protocol based on activation covariance and per-sample gradient SVD spectra. This dual-view reveals three empirical findings and one mechanistic explanation. First, batch size acts as a latent determinant of representation geometry: runs that reach equal loss settle into systematically distinct activation spectra. Second, the activation covariance tail measured early in training reliably forecasts downstream token efficiency. Third, movement of the activation spectrum head (leading modes), together with gradient spectra, characterizes underlying learning-dynamics changes, separating learning-side architectural improvements from primarily execution-side gains. These predictive and diagnostic signals persist across the 12-, 36-, and 48-layer model tiers. Finally, a mechanistic model proves the main observations and explains how activation covariance spectra correlate with task-aligned feature learning.
2026-05-06
arXiv (preprint)
doi.org
arxiv.org