Publications

Simulation-free Structure Learning for Stochastic Population Dynamics

Stephen Y. Zhang

Modeling dynamical systems and unraveling their underlying structural dependencies is central to many domains in the natural sciences. Vario… (voir plus)us physical systems, such as those arising in cell biology, are inherently high-dimensional and stochastic in nature, and admit only partial, noisy state measurements. Our primary motivating setting is single-cell biology, where destructive measurements yield unpaired population snapshots rather than longitudinal trajectories of the same cells. This poses a significant challenge for addressing the problems of modeling the underlying dynamics and inferring the network structure of these systems. Existing methods are typically tailored either for structure learning or modeling dynamics at the population level, but are limited in their ability to address both problems together. In this work, we address both problems simultaneously: we present StructureFlow, a novel and principled simulation-free training approach for jointly learning the structure and stochastic population dynamics of physical systems. We showcase the utility of StructureFlow for the tasks of structure learning from interventions and dynamical (trajectory) inference of conditional population dynamics. We empirically evaluate our approach on high-dimensional synthetic systems, a set of biologically plausible simulated systems, and an experimental single-cell dataset. We show that StructureFlow can learn the structure of underlying systems while simultaneously modeling their conditional population dynamics --- a key step toward model-based mechanistic understanding of systems behavior.

2026-08-30

Transactions on Machine Learning Research (accepté)

openreview.net

Temporal coupling as a design principle for social AI

Annemarie Wolff

Vincent Chamberland

Despite their ability to emulate social reasoning, current AI systems consistently fail at the dynamic, real-time coordination that defines … (voir plus)human social interaction. We argue this failure reflects a missing aspect in their design, not a limitation of scale or training data. Current AI systems model their partner from the outside and train on the data from interactions. Social intelligence, however, is not a property of an individual. It emerges when two coupled individuals interact and is developed through that coupling. We here argue that social AI should be built, and evaluated, as a coupled system with a human which targets an optimal, rather than maximal, alignment. This argument is based on two principles from research in social neuroscience. First, the training objective must change from predicting their interacting partner to a directed coupling that optimizes synchrony. Second, optimizing synchrony requires an embodied and organized temporal architecture which separates faster interactive generation with slower contextual comprehension. We show how coupling can be measured against benchmarks of human interaction and argue that this framing is a safety requirement. Together, these principles offer a neurobiologically grounded pathway toward AI systems capable of real-time social coordination with human partners.

2026-08-30

Zenodo (prépublication)

TPR-Attention for Combinatorial Generalization

Melisa Civelekoğlu

Isabeau Prémont-Schwarz

Systematic generalization remains a significant challenge in deep learning. In particular, combinatorial generalization - generalizing to ne… (voir plus)w configurations of known factors of variation - is effortless for humans but difficult for standard neural architectures that rely on statistical correlations rather than explicit structural representations. We introduce a new architectural component that embeds structured inductive bias into deep learning: an attention mechanism operating over tensor-product representations (TPRs). Through controlled experiments on compositional tasks, we show that this TPR-attention mechanism outperforms existing architectural components in combinatorial generalization. These results highlight the value of integrating explicit compositional structure into neural attention and point toward a promising path for models capable of systematic generalization.

2026-08-30

arXiv (prépublication)

Beyond Document Grounding: Span-Level Hallucination Detection over Code, Tool Output, and Documents

Ádám Kovács

Bowei He

Xue Liu

István Boros

Szilveszter Tóth

Gábor Recski

Hallucination detection for retrieval-augmented generation (RAG) is usually evaluated on natural-language document evidence. However, ground… (voir plus)ed generation systems increasingly rely on structured inputs: source code, developer-tool output, markdown documents, tables, and repository metadata. We introduce a unified benchmark for span-level hallucination detection over code, tool output, structured documents, and existing natural-language RAG datasets. The benchmark is built by starting from grounded correct answers, injecting localized hallucinations with exact character labels, and validating the code test split with evidence-based review. Our fine-tuned Qwen3.5-2B detector reaches 0.689 span-F1 on the unified test set and 0.60 on the code-agent source, where it substantially outperforms LettuceDetect-large (0.17) and the strongest zero-shot LLM judges we evaluated (at most 0.22). The same model remains competitive on established natural-language benchmarks, with 81.8 RAGTruth example-F1 and 0.724 English PsiloQA IoU.

2026-08-29

GroundLM @ Conference on Empirical Methods in Natural Language Processing (publié)

openreview.net

Outcomes of patients with Waldenström macroglobulinemia in ICU: a retrospective multicentric study

Camille Barraud

Aalliayah Velletrie

Stéphanie Harel

Frédéric Pène

Julien Charpentier

Antoine Guillon

Elie Azoulay

Michael Darmon

Emmanuel Canet

Anne-Sophie Moreau

Lucie Biard

Lara Zafrani

BACKGROUND Waldenström macroglobulinemia (WM) is a rare lymphoplasmacytic lymphoma, that can lead to life-threatening complications requiri… (voir plus)ng intensive care unit (ICU) admission. This study aimed to describe the clinical characteristics, WM-specific complications, and outcomes of WM patients admitted to the ICU. METHODS We conducted a multicenter retrospective study in 6 French ICUs including all patients with WM admitted between 2010 and 2024. The primary outcome was ICU mortality, analyzed using Ridge logistic regression. One-year mortality and associated risk factors were also assessed after ICU admission. RESULTS A total of 151 patients with WM were admitted to the ICU over the 14-year period. ICU mortality was 20% (n = 30), with infections being the leading cause of death. The main reason for admission was acute respiratory failure, and nearly half of the patients presented with sepsis at admission. The most WM-specific complications included hyperviscosity syndrome (n = 29, 19%) and other IgM-related vascular complications (n = 27, 18%). In Ridge logistic regression, a higher SOFA score (β = 0.099) and autoimmune haemolytic anemia (β = 0.058) were most strongly associated with ICU mortality, whereas, hyperviscosity syndrome (β = -0.045) showed the strongest negative association. The one-year mortality rate was 40% (n = 58) and remission at one year was achieved in 47% (n = 56) of treated patients. CONCLUSION This is the first large retrospective cohort of critically ill patients with WM. SOFA score and autoimmune cytopenias were most strongly associated with ICU mortality, highlighting the importance of both disease-specific complications and baseline severity, and supporting the need for tailored prognostic tools and individualized management strategies.

2026-08-28

Journal of Critical Care (publié)

Sparse Koopman Autoencoders Identify Local Dynamical Regimes in Multibasin Systems

Aidan Li

Uday Kiran Reddy Tadipatri

Mahan Fathi

Sarath Chandar

Ross Goroshin

Koopman autoencoders (KAEs) seek a higher-dimensional latent representation in which nonlinear dynamics evolve linearly. However, many inter… (voir plus)esting systems have multiple basins of attraction, and both theoretical and empirical work has shown these multibasin systems cannot generally admit a single finite-dimensional global Koopman embedding under standard assumptions. We posit that encoders with a sparsity-inducing objective encouraging few active latent coefficients will provide latent supports as an inspectable basin-modeling principle for Koopman autoencoders. We use these encoders producing sparse latents in training Sparse Koopman Autoencoders (SKAEs) without basin labels or other regime annotations, and treat the learned latent supports as model-produced regime variables after training. Across a range of procedurally generated multibasin systems and chaotic flows, we show that SKAEs have superior forecasting performance compared to dense-latent KAEs. We also perform a mechanistic study that shows latent supports produced by SKAEs are both essential for the quality of the representation and useful for identifying basins on held-out basin interior states, whereas dense-latent KAEs collapse to an uninformative single family. These results identify sparse latents and their corresponding supports as label-free, interpretable regime variables for Koopman learning in nonlinear systems with multiple local dynamical laws.

2026-08-28

arXiv (prépublication)

Emotion reconfigures neuronal coordination across brains to control social choice

Francesco Papaleo

Federica Antonelli

Anna Monai

Fabrizio Bernardi

Atesh Koul

Nicola Marie Engel

Ilaria Carta

Francesca Managò

Arianna Benedetti

Anna Maria Borruto

LUDOVICO SPATTINI

Federica Maltese

Chunzhi Yi

Gian Marco Leggio

Stefano Panzeri

Giacomo Novembre

2026-08-27

Research Square (publié)

Humans and LLMs Diverge on Probabilistic Inferences

Gaurav Kamath

Sreenath Madathil

Sebastian Schuster

Marie-Catherine de Marneffe

Siva Reddy

Abstract Human reasoning often involves working over limited information to arrive at probabilistic conclusions. In its simplest form, this … (voir plus)involves making an inference that is not strictly entailed by a premise, but rather only likely given the premise. While reasoning LLMs have demonstrated strong performance on logical and mathematical tasks, their behavior on such open-ended, non-deterministic inferences remains largely unexplored. We introduce PROBCOPA, a dataset of 210 handcrafted probabilistic inferences in English, each annotated for inference likelihood by 25–30 human participants. We find that human responses are graded and varied, revealing probabilistic judgments of the inferences in our dataset. Comparing these judgments with responses from eight state-of-the-art reasoning LLMs, we show that models consistently fail to produce human-like distributions. Finally, analyzing LLM reasoning chains, we find evidence of a common reasoning pattern used to evaluate such inferences. Our findings reveal persistent differences between humans and LLMs, and underscore the need to evaluate reasoning beyond deterministic settings.1

2026-08-27

Transactions of the Association for Computational Linguistics (publié)

Circuit Discovery Helps Detect LLM Jailbreaking: A Mechanistic Interpretability Study

Despite extensive safety alignment, large language models (LLMs) remain vulnerable to jailbreak attacks that bypass safeguards to elicit har… (voir plus)mful content. While prior work attributes this vulnerability to safety training limitations, the internal mechanisms by which LLMs process adversarial prompts remain poorly understood. We present a mechanistic analysis of the jailbreaking behavior in a large-scale, safety-aligned LLM, focusing on LLaMA-2-7B-chat-hf. Leveraging edge attribution patching and subnetwork probing, we systematically identify computational circuits responsible for generating affirmative responses to jailbreak prompts. Ablating these circuits during the first token prediction can reduce attack success rates by up to 80\%, demonstrating its critical role in safety bypass. Our analysis uncovers key attention heads and MLP pathways that mediate adversarial prompt exploitation, revealing how important tokens propagate through these components to override safety constraints. These findings advance the understanding of adversarial vulnerabilities in aligned LLMs and pave the way for targeted, interpretable defense mechanisms based on mechanistic interpretability.

2026-08-26

arXiv (prépublication)

Green light exposure reduces central sensitization in cats with osteoarthritis-associated somatosensory impairment

Aliénor Delsart

Aude Castel

Frédéric Sauvé

Colombe Otis

Bertrand Lussier

Franck Péron

Marc Hébert

Nicolas Lapointe

Nicolas Quaegebeur

Pierre Grandjean

Maxim Moreau

Johanne Martel-Pelletier

Jean‐Pierre Pelletier

Eric Troncy

In humans, chronic pain is associated with multisensory hypersensitivity, which can be modulated by sensorial exposure. Although cats experi… (voir plus)ence chronic pain, their sensorial perception under this condition has never been investigated. Adult neutered healthy ( n = 6) or osteoarthritic ( n = 32) cats were assessed for functional alteration, central sensitization, and sensorial (olfactory and auditory) tolerance. The effects of specific light wavelengths were tested for neuromodulation via visual exposure. Hypersensitized cats displayed stronger unpleasant odor avoidance and lower sensitivity to repeated repulsive sounds. Odor avoidance correlated with delta and theta band proportions, inverse markers of endogenous inhibitory control. Functional alteration correlated with beta activity, a pain marker. After green light exposure, neuro-sensitization decreased for 80% of the less sensitized cats, and delta and theta band proportions increased further (no longer requiring inhibition), while beta and gamma band proportions decreased. Neuro-sensitization appears central in feline osteoarthritis pain, is associated with sensorial modifications, and is modulable by non-pharmacological approaches.

2026-08-26

iScience (publié)

Top-down feedback in deep neural networks leads to functional differences during audiovisual integration

Artificial neural networks (ANNs) are an important tool for studying neural computation, but many features of the brain are not captured by … (voir plus)standard ANN architectures. One notable missing feature in most ANN models is top-down feedback, that is projections from higher-order layers to lower-order layers in the network. Top-down feedback is ubiquitous in the brain, and it has a unique modulatory impact on activity in neocortical pyramidal neurons. However, we still do not understand its computational role. Here, we develop a deep neural network model that captures the core functional properties of top-down feedback in the neocortex, allowing us to construct hierarchical recurrent ANN models that more closely reflect the architecture of the brain. We use this to explore the impact of different hierarchical recurrent architectures on an audiovisual integration task. We find that certain hierarchies, namely those that mimic the architecture of the human brain, impart ANN models with a light visual bias similar to that seen in humans. This bias does not impair performance on the audiovisual tasks. The results further suggest that different configurations of top-down feedback make otherwise identically connected models functionally distinct from each other, and from traditional feedforward and laterally recurrent models. Altogether, our findings demonstrate that modulatory top-down feedback is a computationally relevant feature of biological brains, and that incorporating it into ANNs affects their behavior and constrains the solutions it is likely to discover.

2026-08-25

eLife (publié)