Search

This paper proposes a proof of concept application of a biphasic constitutive model to identify the mechanical properties of in vivo human skin under extension. Although poromechanics theory has been extensively used to model other soft biological tissues, only a few studies have been published for skin, and most have been limited to ex vivo or in silico conditions. However, in vivo procedures are crucial to determine the subject-specific properties at different body sites. This study focuses on cyclic uni-axial extension of the upper arm skin, using unpublished data collected by Chambert et al. Our analysis shows that a two-layer finite element model allows representing all relevant features of the observed mechanical response to the imposed external loading, which was composed, in this contribution, of four loading-sustaining-unloading cycles. The Root Mean Square Error (RMSE) between the calibrated model and the measured Force-time response was 8.84e-3 N. Our biphasic model represents a preliminary step toward investigating the mechanical conditions responsible for the onset of injury. It allows for the analysis of changes in Interstitial Fluid (IF) pressure, flow, and osmotic pressure, in addition to the mechanical fields. Future work will focus on the interaction of multiple biochemical factors and the complex network of regulatory signals.

A parametric study is conducted to quantify the effect of the keeper electrode geometry on the neutral flow quantities within orificed hollow cathodes, prior to cathode ignition. The keeper impinges directly on the flow out of the cathode orifice and its geometry strongly influences the product between the pressure in the orifice-keeper region and the distance between cathode and keeper, $P_{ko}\cdot D_{ko}$, a key parameter for successful cathode ignition. A representative cathode equipped with a keeper is simulated using the Direct Simulation Monte Carlo method. The numerical model is first validated with computational results from the literature, and a parametric study is then conducted. Parameters include the cathode pressure-diameter in the range of 1-5 Torr-cm and the following geometric ratios: cathode orifice-to-cathode inner radii in the range of 0.1-0.7, keeper orifice-to-cathode orifice radii in the range of 1-5, and keeper distance-to-cathode-orifice diameter, in the range of 0.5-10. If both keeper and cathode have identical orifice radii, the flow remains subsonic in the cathode orifice-to-keeper region. In most cases, however, the flow becomes underexpanded and supersonic, and the static pressure within the cathode orifice-keeper region is, on average, 4% that of the upstream pressure value. The orifice-keeper region pressure increases with either a decrease in the keeper orifice diameter or an increase in the distance between the cathode orifice plate and the keeper plate, in agreement with literature data. Trends are explained through control-volume-based conservation laws. The ratio of ignition-to-nominal mass flow rates is found to be in the range of 9-120, with a most probable value of 40, in agreement with literature data. This suggests that heater-less cathode ignition at a minimum DC voltage may be achieved by increasing the input mass flow rate by a factor of 40., Comment: 25 pages, 27 figures, 4 tables. Submitted to Journal of Applied Physics

The automatic generation of hints by Large Language Models (LLMs) within Intelligent Tutoring Systems (ITSs) has shown potential to enhance student learning. However, generating pedagogically sound hints that address student misconceptions and adhere to specific educational objectives remains challenging. This work explores using LLMs (GPT-4o and Llama-3-8B-instruct) as teachers to generate effective hints for students simulated through LLMs (GPT-3.5-turbo, Llama-3-8B-Instruct, or Mistral-7B-instruct-v0.3) tackling math exercises designed for human high-school students, and designed using cognitive science principles. We present here the study of several dimensions: 1) identifying error patterns made by simulated students on secondary-level math exercises; 2) developing various prompts for GPT-4o as a teacher and evaluating their effectiveness in generating hints that enable simulated students to self-correct; and 3) testing the best-performing prompts, based on their ability to produce relevant hints and facilitate error correction, with Llama-3-8B-Instruct as the teacher, allowing for a performance comparison with GPT-4o. The results show that model errors increase with higher temperature settings. Notably, when hints are generated by GPT-4o, the most effective prompts include prompts tailored to specific errors as well as prompts providing general hints based on common mathematical errors. Interestingly, Llama-3-8B-Instruct as a teacher showed better overall performance than GPT-4o. Also the problem-solving and response revision capabilities of the LLMs as students, particularly GPT-3.5-turbo, improved significantly after receiving hints, especially at lower temperature settings. However, models like Mistral-7B-Instruct demonstrated a decline in performance as the temperature increased., Comment: Accepted at NeurIPS 2024 Workshop on Large Foundation Models for Educational Assessment (FM-Assess)

Reinforcement learning (RL) is a promising approach for aligning large language models (LLMs) knowledge with sequential decision-making tasks. However, few studies have thoroughly investigated the impact on LLM agents capabilities of fine-tuning them with RL in a specific environment. In this paper, we propose a novel framework to analyze the sensitivity of LLMs to prompt formulations following RL training in a textual environment. Our findings reveal that the performance of LLMs degrades when faced with prompt formulations different from those used during the RL training phase. Besides, we analyze the source of this sensitivity by examining the model's internal representations and salient tokens. Finally, we propose to use a contrastive loss to mitigate this sensitivity and improve the robustness and generalization capabilities of LLMs.

The past years have seen Large Language Models (LLMs) strive not only as generative models but also as agents solving textual sequential decision-making tasks. When facing complex environments where their zero-shot abilities are insufficient, recent work showed online Reinforcement Learning (RL) could be used for the LLM agent to discover and learn efficient strategies interactively. However, most prior work sticks to on-policy algorithms, which greatly reduces the scope of methods such agents could use for both exploration and exploitation, such as experience replay and hindsight relabeling. Yet, such methods may be key for LLM learning agents, and in particular when designing autonomous intrinsically motivated agents sampling and pursuing their own goals (i.e. autotelic agents). This paper presents and studies an adaptation of Soft Actor-Critic and hindsight relabeling to LLM agents. Our method not only paves the path towards autotelic LLM agents that learn online but can also outperform on-policy methods in more classic multi-goal RL environments.

Three-dimensional reconstruction is a fundamental problem in robotics perception. We examine the problem of active view selection to perform 3D Gaussian Splatting reconstructions with as few input images as possible. Although 3D Gaussian Splatting has made significant progress in image rendering and 3D reconstruction, the quality of the reconstruction is strongly impacted by the selection of 2D images and the estimation of camera poses through Structure-from-Motion (SfM) algorithms. Current methods to select views that rely on uncertainties from occlusions, depth ambiguities, or neural network predictions directly are insufficient to handle the issue and struggle to generalize to new scenes. By ranking the potential views in the frequency domain, we are able to effectively estimate the potential information gain of new viewpoints without ground truth data. By overcoming current constraints on model architecture and efficacy, our method achieves state-of-the-art results in view selection, demonstrating its potential for efficient image-based 3D reconstruction., Comment: 8 pages, 4 figures

Over the past decade, Magnetic Resonance Fingerprinting (MRF) has emerged as an efficient paradigm for the rapid and simultaneous quantification of multiple MRI parameters, including fat fraction (FF), water T1 ($T1_{H2O}$), water T2 ($T2_{H2O}$), and fat T1 ($T1_{fat}$). These parameters serve as promising imaging biomarkers in various anatomical targets such as the heart, liver, and skeletal muscles. However, measuring these parameters in the upper body poses challenges due to physiological motion, particularly respiratory motion. In this work, we propose a novel approach, motion-corrected (MoCo) MRF T1-FF, which estimates the motion field using an optimized preliminary motion scan and uses it to correct the MRF acquisition data before dictionary search for reconstructing motion-corrected FF and $T1_{H2O}$ parametric maps of the upper-body region. We validated this framework using an $\textit{in vivo}$ dataset comprising ten healthy volunteers and a 10-year-old boy with Duchenne muscular dystrophy. At the ROI level, in regions minimally affected by motion, no significant bias was observed between the uncorrected and MoCo reconstructions for FF (mean difference of -0.7%) and $T1_{H2O}$ (-4.9 ms) values. Moreover, MoCo MRF T1-FF significantly reduced the standard deviations of distributions assessed in these regions, indicating improved precision. Notably, in regions heavily affected by motion, such as respiratory muscles, liver, and kidneys, the MRF parametric maps exhibited a marked reduction in motion blurring and streaking artifacts after motion correction. Furthermore, the diaphragm was consistently discernible on parametric maps after motion correction. This approach lays the groundwork for the joint 3D quantification of FF and $T1_{H2O}$ in regions that are rarely studied, such as the respiratory muscles, particularly the intercostal muscles and diaphragm., Comment: 19 pages, 9 figures, 3 tables

In machine learning, contamination refers to situations where testing data leak into the training set. The issue is particularly relevant for the evaluation of the performance of Large Language Models (LLMs), which are generally trained on gargantuan, and generally opaque, corpora of text scraped from the world wide web. Developing tools to detect contamination is therefore crucial to be able to fairly and properly track the evolution of the performance of LLMs. Most recent works in the field are not tailored to quantify contamination on short sequences of text like we find in psychology questionnaires. In the present paper we introduce LogProber, a novel, efficient, algorithm that we show able to detect contamination using token probability in given sentences. In the second part we investigate the limitations of the method and discuss how different training methods can contaminate models without leaving traces in the token probabilities.

Through basin-scale circulations, the ocean regulates global distributions of heat, nutrients, and greenhouse gases. To properly predict the future of the ocean under climate change, we need to develop a thorough understanding of the underlying mechanisms that drive global circulations. An estimated 2 TW of power is required to support interior mixing. Roughly half of this power is believed to come from tidal flow over topography, producing internal gravity waves (IGW's), which can radiate energy throughout the ocean interior. But it is difficult to track the subsequent journey from tidal injection to dissipation, as the energy cascade spans an enormous range of spatio-temporal scales and multiple different nonlinear transfer mechanisms. To investigate the full energy pathway from topographic forcing to irreversible mixing, we built a model ocean in a large-scale laboratory wavetank (9 m x 2.9 m x 0.75 m) allowing Reynolds numbers up to O(10$^5$). We replicate the tidal forcing by oscillating an idealized ocean ridge. We track energy transfer across the first cascade, driven by wave turbulence, using Background Oriented Schlieren (BOS) over the full tank. Through the BOS we observe the formation of various sets of subharmonics, driven by Triadic Resonant Instabilities (TRI). At later times, the subharmonics born from TRI engage in different interactions, which ultimately develop a continuum of waves at frequencies up to $N$. We validate the three-wave resonant conditions through a Fourier decomposition and confirm a backward cascade in frequency but a forward cascade in vertical wavenumber. Through our spatial analysis, we identify relevant three-wave interactions and show the significance of elastic scattering, a nonlocal interaction, in our fully evolved system. We note however that the majority of our triads are local, which have been historically overlooked., Comment: 21 pages, 15 figures

Representing and quantifying Minimal Residual Disease (MRD) in Acute Myeloid Leukemia (AML), a type of cancer that affects the blood and bone marrow, is essential in the prognosis and follow-up of AML patients. As traditional cytological analysis cannot detect leukemia cells below 5\%, the analysis of flow cytometry dataset is expected to provide more reliable results. In this paper, we explore statistical learning methods based on optimal transport (OT) to achieve a relevant low-dimensional representation of multi-patient flow cytometry measurements (FCM) datasets considered as high-dimensional probability distributions. Using the framework of OT, we justify the use of the K-means algorithm for dimensionality reduction of multiple large-scale point clouds through mean measure quantization by merging all the data into a single point cloud. After this quantization step, the visualization of the intra and inter-patients FCM variability is carried out by embedding low-dimensional quantized probability measures into a linear space using either Wasserstein Principal Component Analysis (PCA) through linearized OT or log-ratio PCA of compositional data. Using a publicly available FCM dataset and a FCM dataset from Bordeaux University Hospital, we demonstrate the benefits of our approach over the popular kernel mean embedding technique for statistical learning from multiple high-dimensional probability distributions. We also highlight the usefulness of our methodology for low-dimensional projection and clustering patient measurements according to their level of MRD in AML from FCM. In particular, our OT-based approach allows a relevant and informative two-dimensional representation of the results of the FlowSom algorithm, a state-of-the-art method for the detection of MRD in AML using multi-patient FCM.

Human culture relies on collective innovation: our ability to continuously explore how existing elements in our environment can be combined to create new ones. Language is hypothesized to play a key role in human culture, driving individual cognitive capacities and shaping communication. Yet the majority of models of collective innovation assign no cognitive capacities or language abilities to agents. Here, we contribute a computational study of collective innovation where agents are Large Language Models (LLMs) that play Little Alchemy 2, a creative video game originally developed for humans that, as we argue, captures useful aspects of innovation landscapes not present in previous test-beds. We, first, study an LLM in isolation and discover that it exhibits both useful skills and crucial limitations. We, then, study groups of LLMs that share information related to their behaviour and focus on the effect of social connectivity on collective performance. In agreement with previous human and computational studies, we observe that groups with dynamic connectivity out-compete fully-connected groups. Our work reveals opportunities and challenges for future studies of collective innovation that are becoming increasingly relevant as Generative Artificial Intelligence algorithms and humans innovate alongside each other.

This work introduces a partitioned optimization framework (POf) to ease the solving process for optimization problems for which fixing some variables to a tunable value removes some difficulties. The variables space may be continuously partitioned into subsets where these variables are fixed to a given value, so that minimizing the objective function restricted to any of the partition sets is easier than minimizing the objective function over the whole variables space. Then the major difficulty is translated from solving the original problem to choosing the partition set minimizing the minimum of the objective function restricted to this set. Hence, a local solution to the original problem is given by computing this partition set and selecting the minimizer of the objective function restricted to this set. This work formalizes this framework, studies its theoretical guarantees, and provides a numerical method to seek for an optimal partition set using a derivative-free algorithm. Some illustrative problems are presented to show how to apply the POf and to highlight the gain in numerical performance it provides.

As large language models (LLMs) start interacting with each other and generating an increasing amount of text online, it becomes crucial to better understand how information is transformed as it passes from one LLM to the next. While significant research has examined individual LLM behaviors, existing studies have largely overlooked the collective behaviors and information distortions arising from iterated LLM interactions. Small biases, negligible at the single output level, risk being amplified in iterated interactions, potentially leading the content to evolve towards attractor states. In a series of telephone game experiments, we apply a transmission chain design borrowed from the human cultural evolution literature: LLM agents iteratively receive, produce, and transmit texts from the previous to the next agent in the chain. By tracking the evolution of text toxicity, positivity, difficulty, and length across transmission chains, we uncover the existence of biases and attractors, and study their dependence on the initial text, the instructions, language model, and model size. For instance, we find that more open-ended instructions lead to stronger attraction effects compared to more constrained tasks. We also find that different text properties display different sensitivity to attraction effects, with toxicity leading to stronger attractors than length. These findings highlight the importance of accounting for multi-step transmission dynamics and represent a first step towards a more comprehensive understanding of LLM cultural dynamics., Comment: Code available at https://github.com/jeremyperez2/TelephoneGameLLM. Companion website with a Data Explorer tool at https://sites.google.com/view/telephone-game-llm

We introduce the LiLiuM series of large language models (LLMs): 1B, 7B, and 13B parameter models developed 100% in-house to fit eBay's specific needs in the e-commerce domain. This gives eBay full control over all aspects of the models including license, data, vocabulary, and architecture. We expect these models to be used as a foundation for fine-tuning and instruction-tuning, eliminating dependencies to external models. The LiLiuM LLMs have been trained on 3 trillion tokens of multilingual text from general and e-commerce domain. They perform similar to the popular LLaMA-2 models on English natural language understanding (NLU) benchmarks. At the same time, we outperform LLaMA-2 on non-English NLU tasks, machine translation and on e-commerce specific downstream tasks. As part of our data mixture, we utilize the newly released RedPajama-V2 dataset for training and share our insights regarding data filtering and deduplication. We also discuss in detail how to serialize structured data for use in autoregressive language modeling. We provide insights on the effects of including code and parallel machine translation data in pre-training. Furthermore, we develop our own tokenizer and model vocabulary, customized towards e-commerce. This way, we can achieve up to 34% speed-up in text generation on eBay-specific downstream tasks compared to LLaMA-2. Finally, in relation to LLM pretraining, we show that checkpoint averaging can further improve over the best individual model checkpoint.

In the trend of hybrid Artificial Intelligence techniques, Physical-Informed Machine Learning has seen a growing interest. It operates mainly by imposing data, learning, or architecture bias with simulation data, Partial Differential Equations, or equivariance and invariance properties. While it has shown great success on tasks involving one physical domain, such as fluid dynamics, existing methods are not adapted to tasks with complex multi-physical and multi-domain phenomena. In addition, it is mainly formulated as an end-to-end learning scheme. To address these challenges, we propose to leverage Bond Graphs, a multi-physics modeling approach, together with Message Passing Graph Neural Networks. We propose a Neural Bond graph Encoder (NBgE) producing multi-physics-informed representations that can be fed into any task-specific model. It provides a unified way to integrate both data and architecture biases in deep learning. Our experiments on two challenging multi-domain physical systems - a Direct Current Motor and the Respiratory System - demonstrate the effectiveness of our approach on a multivariate time-series forecasting task., Comment: 9 pages, 3 figures, paper under review

We study the semigroups of random Schr\"odinger operators of the form $\widehat{H}f=-\frac12f''+(V+\xi)f$, where $f:I\to\mathbb F^r$ ($\mathbb F=\mathbb R,\mathbb C,\mathbb H$) are vector-valued functions on a possibly infinite interval $I\subset\mathbb R$ that satisfy a mix of Robin and Dirichlet boundary conditions, $V$ is a deterministic diagonal potential with power-law growth at infinity, and $\xi$ is a matrix white noise. Our main result consists of Feynman-Kac formulas for trace moments of the form $\mathbf E[\prod_{k=1}^n\mathrm{Tr}[\mathrm e^{-t_k\widehat{H}}]]$ ($n\in\mathbb N$, $t_k>0$). One notable example covered by our main result consists of the multivariate stochastic Airy operator (SAO) of Bloemendal and Vir\'ag (Ann. Probab., 44(4):2726-2769, 2016), which characterizes the soft-edge eigenvalue fluctuations of critical rank-$r$ spiked Wishart and GO/U/SE random matrices. As a corollary of our main result, we prove that if $V$'s growth is at least linear (this includes the multivariate SAO), then $\widehat{H}$'s spectrum is number rigid in the sense of Ghosh and Peres (Duke Math. J., 166(10):1789-1858, 2017). Together with the rigidity of the scalar SAO, this completes the characterization of number rigidity in the soft-edge limits of Gaussian $\beta$-ensembles and their finite-rank spiked versions., Comment: 90 pages, 8 figures. This preprint supersedes arXiv:2311.08564, along with some important bibliographical updates

This paper addresses the problem of pathological lung segmentation, a significant challenge in medical image analysis, particularly pronounced in cases of peripheral opacities (severe fibrosis and consolidation) because of the textural similarity between lung tissue and surrounding areas. To overcome these challenges, this paper emphasizes the use of CycleGAN for unpaired image-to-image translation, in order to provide an augmentation method able to generate fake pathological images matching an existing ground truth. Although previous studies have employed CycleGAN, they often neglect the challenge of shape deformation, which is crucial for accurate medical image segmentation. Our work introduces an innovative strategy that incorporates additional loss functions. Specifically, it proposes an L1 loss based on the lung surrounding which shape is constrained to remain unchanged at the transition from the healthy to pathological domains. The lung surrounding is derived based on ground truth lung masks available in the healthy domain. Furthermore, preprocessing steps, such as cropping based on ribs/vertebra locations, are applied to refine the input for the CycleGAN, ensuring that the network focus on the lung region. This is essential to avoid extraneous biases, such as the zoom effect bias, which can divert attention from the main task. The method is applied to enhance in semi-supervised manner the lung segmentation process by employing a U-Net model trained with on-the-fly data augmentation incorporating synthetic pathological tissues generated by the CycleGAN model. Preliminary results from this research demonstrate significant qualitative and quantitative improvements, setting a new benchmark in the field of pathological lung segmentation. Our code is available at https://github.com/noureddinekhiati/Semi-supervised-lung-segmentation, Comment: 14 pages, 7 figures

The emerging behaviors of swarms have fascinated scientists and gathered significant interest in the field of robotics. Traditionally, swarms are viewed as egalitarian, with robots sharing identical roles and capabilities. However, recent findings highlight the importance of hierarchy for deploying robot swarms more effectively in diverse scenarios. Despite nature's preference for hierarchies, the robotics field has clung to the egalitarian model, partly due to a lack of empirical evidence for the conditions favoring hierarchies. Our research demonstrates that while egalitarian swarms excel in environments proportionate to their collective sensing abilities, they struggle in larger or more complex settings. Hierarchical swarms, conversely, extend their sensing reach efficiently, proving successful in larger, more unstructured environments with fewer resources. We validated these concepts through simulations and physical robot experiments, using a complex radiation cleanup task. This study paves the way for developing adaptable, hierarchical swarm systems applicable in areas like planetary exploration and autonomous vehicles. Moreover, these insights could deepen our understanding of hierarchical structures in biological organisms., Comment: 31 Pages, 7 Figures. Supplementary material attached to the paper

The performance of Automated Recognition (ATR) algorithms on side-scan sonar imagery has shown to degrade rapidly when deployed on non benign environments. Complex seafloors and acoustic artefacts constitute distractors in the form of strong textural patterns, creating false detections or preventing detections of true objects. This paper presents two online seafloor characterisation techniques to improve explainability during Autonomous Underwater Vehicles (AUVs) missions. Importantly and as opposed to previous work in the domain, these techniques are not based on a model and require limited input from human operators, making it suitable for real-time onboard processing. Both techniques rely on an unsupervised machine learning approach to extract terrain features which relate to the human understanding of terrain complexity. The first technnique provides a quantitative, application-driven terrain characterisation metric based on the performance of an ATR algorithm. The second method provides a way to incorporate subject matter expertise and enables contextualisation and explainability in support for scenario-dependent subjective terrain characterisation. The terrain complexity matches the expectation of seasoned users making this tool desirable and trustworthy in comparison to traditional unsupervised approaches. We finally detail an application of these techniques to repair a Mine Countermeasures (MCM) mission carried with SeeByte autonomy framework Neptune., Comment: Presented at UACE (Underwater Acoustics Conference & Exhibition) 2023, Kalamata, Greece

This paper introduces PhyloLM, a method adapting phylogenetic algorithms to Large Language Models (LLMs) to explore whether and how they relate to each other and to predict their performance characteristics. Our method calculates a phylogenetic distance metrics based on the similarity of LLMs' output. The resulting metric is then used to construct dendrograms, which satisfactorily capture known relationships across a set of 111 open-source and 45 closed models. Furthermore, our phylogenetic distance predicts performance in standard benchmarks, thus demonstrating its functional validity and paving the way for a time and cost-effective estimation of LLM capabilities. To sum up, by translating population genetic concepts to machine learning, we propose and validate a tool to evaluate LLM development, relationships and capabilities, even in the absence of transparent training information.

This paper presents a novel adversary model specifically tailored to distributed systems, aiming to assess the security of blockchain networks. Building upon concepts such as adversarial assumptions, goals, and capabilities, our proposed adversary model classifies and constrains the use of adversarial actions based on classical distributed system models, defined by both failure and communication models. The objective is to study the effects of these allowed actions on the properties of distributed protocols under various system models. A significant aspect of our research involves integrating this adversary model into the Multi-Agent eXperimenter (MAX) framework. This integration enables fine-grained simulations of adversarial attacks on blockchain networks. In this paper, we particularly study four distinct fairness properties on Hyperledger Fabric with the Byzantine Fault Tolerant Tendermint consensus algorithm being selected for its ordering service. We define novel attacks that combine adversarial actions on both protocols, with the aim of violating a specific client-fairness property. Simulations confirm our ability to violate this property and allow us to evaluate the impact of these attacks on several order-fairness properties that relate orders of transaction reception and delivery., Comment: 10 pages, 8 figures

Intraventricular vector flow mapping (iVFM) seeks to enhance and quantify color Doppler in cardiac imaging. In this study, we propose novel alternatives to the traditional iVFM optimization scheme by utilizing physics-informed neural networks (PINNs) and a physics-guided nnU-Net-based supervised approach. When evaluated on simulated color Doppler images derived from a patient-specific computational fluid dynamics model and in vivo Doppler acquisitions, both approaches demonstrate comparable reconstruction performance to the original iVFM algorithm. The efficiency of PINNs is boosted through dual-stage optimization and pre-optimized weights. On the other hand, the nnU-Net method excels in generalizability and real-time capabilities. Notably, nnU-Net shows superior robustness on sparse and truncated Doppler data while maintaining independence from explicit boundary conditions. Overall, our results highlight the effectiveness of these methods in reconstructing intraventricular vector blood flow. The study also suggests potential applications of PINNs in ultrafast color Doppler imaging and the incorporation of fluid dynamics equations to derive biomarkers for cardiovascular diseases based on blood flow., Comment: 12 pages, accepted for publication in IEEE TUFFC; camera ready corrections, corrected acknowledgments

Curiosity-driven learning has shown significant positive effects on students' learning experiences and outcomes. But despite this importance, reports show that children lack this skill, especially in formal educational settings. To address this challenge, we propose an 8-session workshop that aims to enhance children's curiosity through training a set of specific metacognitive skills we hypothesize are involved in its process. Our workshop contains animated videos presenting declarative knowledge about curiosity and the said metacognitive skills as well as practice sessions to apply these skills during a reading-comprehension task, using a web platform designed for this study (e.g. expressing uncertainty, formulating questions, etc). We conduct a pilot study with 15 primary school students, aged between 8 and 10. Our first results show a positive impact on children's metacognitive efficiency and their ability to express their curiosity through question-asking behaviors.

Research in cultural evolution aims at providing causal explanations for the change of culture over time. Over the past decades, this field has generated an important body of knowledge, using experimental, historical, and computational methods. While computational models have been very successful at generating testable hypotheses about the effects of several factors, such as population structure or transmission biases, some phenomena have so far been more complex to capture using agent-based and formal models. This is in particular the case for the effect of the transformations of social information induced by evolved cognitive mechanisms. We here propose that leveraging the capacity of Large Language Models (LLMs) to mimic human behavior may be fruitful to address this gap. On top of being an useful approximation of human cultural dynamics, multi-agents models featuring generative agents are also important to study for their own sake. Indeed, as artificial agents are bound to participate more and more to the evolution of culture, it is crucial to better understand the dynamics of machine-generated cultural evolution. We here present a framework for simulating cultural evolution in populations of LLMs, allowing the manipulation of variables known to be important in cultural evolution, such as network structure, personality, and the way social information is aggregated and transformed. The software we developed for conducting these simulations is open-source and features an intuitive user-interface, which we hope will help to build bridges between the fields of cultural evolution and generative artificial intelligence., Comment: 17 pages, 20 figures. Open-source code available at https://github.com/jeremyperez2/LLM-Culture

We consider the Stochastic Heat Equation (SHE) in $(1+1)$ dimensions with delta Dirac initial data and spacetime white noise. We prove exact large-time asymptotics for multi-point correlations of the SHE for strictly sublinear space coordinates. The sublinear condition is optimal, in the sense that different asymptotics are known to occur when the space coordinates grow linearly [Lin 2023, Theorem 1.1]. Lastly, a notable feature of our result is that it confirms the connection between multi-point correlations in the SHE and the ground state of the Hamiltonian of the delta-Bose gas., Comment: 10 pages, comments welcome!

The Finite Element Method (FEM) is a well-established procedure for computing approximate solutions to deterministic engineering problems described by partial differential equations. FEM produces discrete approximations of the solution with a discretisation error that can be an be quantified with \emph{a posteriori} error estimates. The practical relevance of error estimates for biomechanics problems, especially for soft tissue where the response is governed by large strains, is rarely addressed. In this contribution, we propose an implementation of \emph{a posteriori} error estimates targeting a user-defined quantity of interest, using the Dual Weighted Residual (DWR) technique tailored to biomechanics. The proposed method considers a general setting that encompasses three-dimensional geometries and model non-linearities, which appear in hyperelastic soft tissues. We take advantage of the automatic differentiation capabilities embedded in modern finite element software, which allows the error estimates to be computed generically for a large class of models and constitutive laws. First we validate our methodology using experimental measurements from silicone samples, and then illustrate its applicability for patient-specific computations of pressure ulcers on a human heel.

The standard way to study Large Language Models (LLMs) with benchmarks or psychology questionnaires is to provide many different queries from similar minimal contexts (e.g. multiple choice questions). However, due to LLMs' highly context-dependent nature, conclusions from such minimal-context evaluations may be little informative about the model's behavior in deployment (where it will be exposed to many new contexts). We argue that context-dependence (specifically, value stability) should be studied as a specific property of LLMs and used as another dimension of LLM comparison (alongside others such as cognitive abilities, knowledge, or model size). We present a case-study on the stability of value expression over different contexts (simulated conversations on different topics) as measured using a standard psychology questionnaire (PVQ) and on behavioral downstream tasks. Reusing methods from psychology, we study Rank-order stability on the population (interpersonal) level, and Ipsative stability on the individual (intrapersonal) level. We consider two settings (with and without instructing LLMs to simulate particular personas), two simulated populations, and three downstream tasks. We observe consistent trends in the stability of models and model families - Mixtral, Mistral, GPT-3.5 and Qwen families are more stable than LLaMa-2 and Phi. The consistency of these trends implies that some models exhibit higher value stability than others, and that stability can be estimated with the set of introduced methodological tools. When instructed to simulate particular personas, LLMs exhibit low Rank-order stability, which further diminishes with conversation length. This highlights the need for future research on LLMs that coherently simulate different personas. This paper provides a foundational step in that direction, and, to our knowledge, it is the first study of value stability in LLMs., Comment: The project website and code are available at https://sites.google.com/view/llmvaluestability Published in PLOS ONE ( https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0309114 ), and a shorter version at CogSci 24 ( https://escholarship.org/uc/item/7w4823c6 )

The research field of Artificial Life studies how life-like phenomena such as autopoiesis, agency, or self-regulation can self-organize in computer simulations. In cellular automata (CA), a key open-question has been whether it it is possible to find environment rules that self-organize robust "individuals" from an initial state with no prior existence of things like "bodies", "brain", "perception" or "action". In this paper, we leverage recent advances in machine learning, combining algorithms for diversity search, curriculum learning and gradient descent, to automate the search of such "individuals", i.e. localized structures that move around with the ability to react in a coherent manner to external obstacles and maintain their integrity, hence primitive forms of sensorimotor agency. We show that this approach enables to find systematically environmental conditions in CA leading to self-organization of such basic forms of agency. Through multiple experiments, we show that the discovered agents have surprisingly robust capabilities to move, maintain their body integrity and navigate among various obstacles. They also show strong generalization abilities, with robustness to changes of scale, random updates or perturbations from the environment not seen during training. We discuss how this approach opens new perspectives in AI and synthetic bioengineering.

RESTful APIs are popular web services, requiring documentation to ease their comprehension, reusability and testing practices. The OpenAPI Specification (OAS) is a widely adopted and machine-readable format used to document such APIs. However, manually documenting RESTful APIs is a time-consuming and error-prone task, resulting in unavailable, incomplete, or imprecise documentation. As RESTful API testing tools require an OpenAPI specification as input, insufficient or informal documentation hampers testing quality. Recently, Large Language Models (LLMs) have demonstrated exceptional abilities to automate tasks based on their colossal training data. Accordingly, such capabilities could be utilized to assist the documentation and testing process of RESTful APIs. In this paper, we present RESTSpecIT, the first automated RESTful API specification inference and black-box testing approach leveraging LLMs. The approach requires minimal user input compared to state-of-the-art RESTful API inference and testing tools; Given an API name and an LLM key, HTTP requests are generated and mutated with data returned by the LLM. By sending the requests to the API endpoint, HTTP responses can be analyzed for inference and testing purposes. RESTSpecIT utilizes an in-context prompt masking strategy, requiring no model fine-tuning. Our evaluation demonstrates that RESTSpecIT is capable of: (1) inferring specifications with 85.05% of GET routes and 81.05% of query parameters found on average, (2) discovering undocumented and valid routes and parameters, and (3) uncovering server errors in RESTful APIs. Inferred specifications can also be used as testing tool inputs.

Large class sizes pose challenges to personalized learning in schools, which educational technologies, especially intelligent tutoring systems (ITS), aim to address. In this context, the ZPDES algorithm, based on the Learning Progress Hypothesis (LPH) and multi-armed bandit machine learning techniques, sequences exercises that maximize learning progress (LP). This algorithm was previously shown in field studies to boost learning performances for a wider diversity of students compared to a hand-designed curriculum. However, its motivational impact was not assessed. Also, ZPDES did not allow students to express choices. This limitation in agency is at odds with the LPH theory concerned with modeling curiosity-driven learning. We here study how the introduction of such choice possibilities impact both learning efficiency and motivation. The given choice concerns dimensions that are orthogonal to exercise difficulty, acting as a playful feature. In an extensive field study (265 7-8 years old children, RCT design), we compare systems based either on ZPDES or a hand-designed curriculum, both with and without self-choice. We first show that ZPDES improves learning performance and produces a positive and motivating learning experience. We then show that the addition of choice triggers intrinsic motivation and reinforces the learning effectiveness of the LP-based personalization. In doing so, it strengthens the links between intrinsic motivation and performance progress during the serious game. Conversely, deleterious effects of the playful feature are observed for hand-designed linear paths. Thus, the intrinsic motivation elicited by a playful feature is beneficial only if the curriculum personalization is effective for the learner. Such a result deserves great attention due to increased use of playful features in non adaptive educational technologies., Comment: 29 pages, 37 figures

Deep learning enables automatic and robust extraction of cardiac function descriptors from echocardiographic sequences, such as ejection fraction or strain. These descriptors provide fine-grained information that physicians consider, in conjunction with more global variables from the clinical record, to assess patients' condition. Drawing on novel transformer models applied to tabular data, we propose a method that considers all descriptors extracted from medical records and echocardiograms to learn the representation of a cardiovascular pathology with a difficult-to-characterize continuum, namely hypertension. Our method first projects each variable into its own representation space using modality-specific approaches. These standardized representations of multimodal data are then fed to a transformer encoder, which learns to merge them into a comprehensive representation of the patient through the task of predicting a clinical rating. This stratification task is formulated as an ordinal classification to enforce a pathological continuum in the representation space. We observe the major trends along this continuum on a cohort of 239 hypertensive patients, providing unprecedented details in the description of hypertension's impact on various cardiac function descriptors. Our analysis shows that i) the XTab foundation model's architecture allows to reach outstanding performance (98% AUROC) even with limited data (less than 200 training samples), ii) stratification across the population is reproducible between trainings (within 3.6% MAE), and iii) patterns emerge in descriptors, some of which align with established physiological knowledge about hypertension, while others could pave the way for a more comprehensive understanding of this pathology., Comment: 12 pages + 2 pages of supplementary material, submitted to IEEE journal

From 31 Earth-based and three Voyager 2 occultations spanning 1977--2006, we determine the orbital elements of the nine main Uranian rings with typical RMS residuals of 0.2 -- 0.4 km and 1-$\sigma$ errors in $a, ae,$ and $a\sin i$ of order 0.1 km, registered on an absolute radius scale accurate to 0.2 km at the 2-$\sigma$ level. The $\lambda$ ring shows more substantial scatter. In addition to the free modes $m=0$ in the $\gamma$ ring and $m=2$ in the $\delta$ ring, we find two additional outer Lindblad resonance (OLR) modes ($m=-1$ and $-2$) and a possible $m=3$ inner Lindblad resonance (ILR) mode in the $\gamma$ ring. No normal modes are detected for rings 6, 5, 4, $\alpha$, or $\beta$. Five normal modes are forced by small moonlets: the 3:2 inner ILR of Cressida with the $\eta$ ring, the 6:5 ILR of Ophelia with the $\gamma$ ring, the 23:22 ILR of Cordelia with the $\delta$ ring, the 14:13 ILR of Ophelia with the outer edge of the $\epsilon$ ring, and the counterpart 25:24 OLR of Cordelia with the ring's inner edge. We determine the width-radius relations for nearly all of the detected mode. We find no convincing evidence for librations of any of the rings. The Uranus pole direction at epoch TDB 1986 Jan 19 12:00 is $\alpha_P=77.311327\pm 0.000141^\circ$ and $\delta_P=15.172795\pm0.000618^\circ$. We determine the zonal gravitational coefficients $J_2=(3509.291\pm0.412)\times 10^{-6}, J_4=(-35.522\pm0.466)\times10^{-6}$, and $J_6$ fixed at $0.5\times 10^{-6}$, with a correlation coefficient $\rho(J_2,J_4)=0.9861$, for a reference radius $R=$25559 km. From the amplitudes and resonance radii of normal modes forced by moonlets, we determine the masses of Cressida, Cordelia, and Ophelia. Their estimated densities decrease systematically with increasing orbital radius and generally follow the radial trend of the Roche critical density for a shape parameter $\gamma=1.6$., Comment: 94 pages, 53 figures