Search Results for author:
Found 74 papers, 35 papers with code
NeurIPS 2025 E2LM Competition : Early Training Evaluation of Language Models
Existing benchmarks have proven effective for assessing the performance of fully trained large language models.
UP-dROM : Uncertainty-Aware and Parametrised dynamic Reduced-Order Model, application to unsteady flows
In this work, we present a nonlinear reduction strategy specifically designed for transient flows that incorporates parametrisation and uncertainty quantification.
SCOPE: A Self-supervised Framework for Improving Faithfulness in Conditional Text Generation
This issue arises in typical conditional text generation tasks, such as text summarization and data-to-text generation, where the goal is to produce fluent text based on contextual input.
GEPS: Boosting Generalization in Parametric PDE Neural Solvers through Adaptive Conditioning
As existing adaptive conditioning methods do not scale well with respect to the number of parameters to adapt in the neural solver, we propose GEPS, a simple adaptation mechanism to boost GEneralization in Pde Solvers via a first-order optimization and low-rank rapid adaptation of a small set of context parameters.
Learning a Neural Solver for Parametric PDE to Enhance Physics-Informed Methods
Physics-informed deep learning often faces optimization challenges due to the complexity of solving partial differential equations (PDEs), which involve exploring large solution spaces, require numerous iterations, and can lead to unstable training.
Probing Language Models on Their Knowledge Source
We evaluate this framework on various LLMs of different sizes and demonstrate that mid-layer activations, particularly those related to relations in the input, are crucial in predicting knowledge source selection, paving the way for more reliable models capable of handling knowledge conflicts effectively.
Zebra: In-Context and Generative Pretraining for Solving Parametric PDEs
Solving time-dependent parametric partial differential equations (PDEs) is challenging, as models must adapt to variations in parameters such as coefficients, forcing terms, and boundary conditions.
MEXMA: Token-level objectives improve sentence representations
Current pre-trained cross-lingual sentence encoders approaches use sentence-level objectives only.
NeurIPS 2024 ML4CFD Competition: Harnessing Machine Learning for Computational Fluid Dynamics in Airfoil Design
The integration of machine learning (ML) techniques for addressing intricate physics problems is increasingly recognized as a promising avenue for expediting simulations.
AROMA: Preserving Spatial Structure for Latent PDE Modeling with Local Neural Fields
We present AROMA (Attentive Reduced Order Model with Attention), a framework designed to enhance the modeling of partial differential equations (PDEs) using local neural fields.
ML4PhySim : Machine Learning for Physical Simulations Challenge (The airfoil design)
The aim of this competition is to encourage the development of new ML techniques to solve physical problems using a unified evaluation framework proposed recently, called Learning Industrial Physical Simulations (LIPS).
LOCOST: State-Space Models for Long Document Abstractive Summarization
State-space models are a low-complexity alternative to transformers for encoding long sequences and capturing long-term dependencies.
INFINITY: Neural Field Modeling for Reynolds-Averaged Navier-Stokes Equations
For numerical design, the development of efficient and accurate surrogate models is paramount.
Operator Learning with Neural Fields: Tackling PDEs on General Geometries
Machine learning approaches for solving partial differential equations require learning mappings between function spaces.
Time Series Continuous Modeling for Imputation and Forecasting with Implicit Neural Representations
We introduce a novel modeling approach for time series imputation and forecasting, tailored to address the challenges often encountered in real-world data, such as irregular samples, missing data, or unaligned measurements from multiple sensors.
Adversarial Sample Detection Through Neural Network Transport Dynamics
We propose a detector of adversarial samples that is based on the view of neural networks as discrete dynamic systems.
Stability of implicit neural networks for long-term forecasting in dynamical systems
We develop a theory based on the stability definition of schemes to ensure the stability in forecasting of this network.
Learning from Multiple Sources for Data-to-Text and Text-to-Data
Our model is designed to handle the tasks of D2T and T2D jointly.
Exemplars and Counterexemplars Explanations for Image Classifiers, Targeting Skin Lesion Labeling
We propose a use case study, for skin lesion diagnosis, illustrating how it is possible to provide the practitioner with explanations on the decisions of a state of the art deep neural network classifier trained to characterize skin lesions from examples.
AirfRANS: High Fidelity Computational Fluid Dynamics Dataset for Approximating Reynolds-Averaged Navier-Stokes Solutions
Surrogate models are necessary to optimize meaningful quantities in physical dynamics as their recursive numerical resolutions are often prohibitively expensive.
HMOE: Hypernetwork-based Mixture of Experts for Domain Generalization
This paper presents a novel DG method, called HMOE: Hypernetwork-based Mixture of Experts (MoE), which does not rely on domain labels and is more interpretable.
Block-wise Training of Residual Networks via the Minimizing Movement Scheme
End-to-end backpropagation has a few shortcomings: it requires loading the entire model during training, which can be impossible in constrained settings, and suffers from three locking problems (forward locking, update locking and backward locking), which prohibit training the layers in parallel.
Continuous PDE Dynamics Forecasting with Implicit Neural Representations
Effective data-driven PDE forecasting methods often rely on fixed spatial and / or temporal discretizations.
Multi-scale Physical Representations for Approximating PDE Solutions with Graph Neural Operators
In this work, we study three multi-resolution schema with integral kernel operators that can be approximated with \emph{Message Passing Graph Neural Networks} (MPGNNs).
An extensible Benchmarking Graph-Mesh dataset for studying Steady-State Incompressible Navier-Stokes Equations
Recent progress in \emph{Geometric Deep Learning} (GDL) has shown its potential to provide powerful data-driven models.
Diverse Weight Averaging for Out-of-Distribution Generalization
Standard neural networks struggle to generalize under distribution shifts in computer vision.
PTFlash : A deep learning framework for isothermal two-phase equilibrium calculations
Phase equilibrium calculations are an essential part of numerical simulations of multi-component multi-phase flow in porous media, accounting for the largest share of the computational time.
Generalizing to New Physical Systems via Context-Informed Dynamics Model
Data-driven approaches to modeling physical systems fail to generalize to unseen systems that share the same general dynamics with the learning domain, but correspond to different physical contexts.
Explainable Deep Image Classifiers for Skin Lesion Diagnosis
A key issue in critical contexts such as medical diagnosis is the interpretability of the deep learning models adopted in decision-making systems.
Mapping conditional distributions for domain adaptation under generalized target shift
We consider the problem of unsupervised domain adaptation (UDA) between a source and a target domain under conditional and label shift a. k. a Generalized Target Shift (GeTarS).
Constrained Physical-Statistics Models for Dynamical System Identification and Prediction
A common practice to identify the respective parameters of the physical and ML components is to formulate the problem as supervised learning on observed trajectories.
Separating Retention from Extraction in the Evaluation of End-to-end Relation Extraction
State-of-the-art NLP models can adopt shallow heuristics that limit their generalization capability (McCoy et al., 2019).
Unsupervised domain adaptation with non-stochastic missing data
Moreover, we compare the target error of our Adaptation-imputation framework and the "ideal" target error of a UDA classifier without missing target components.
Differentiable Feature Selection, a Reparameterization Approach
The second one corresponds to the elimination of redundant information by selecting variables in a correlated fashion which requires modeling the covariance of the binary distribution.
A Neural Tangent Kernel Perspective of GANs
We propose a novel theoretical framework of analysis for Generative Adversarial Networks (GANs).
LEADS: Learning Dynamical Systems that Generalize Across Environments
Both are sub-optimal: the former disregards the discrepancies between environments leading to biased solutions, while the latter does not exploit their potential commonalities and is prone to scarcity problems.
Data-QuestEval: A Referenceless Metric for Data-to-Text Semantic Evaluation
QuestEval is a reference-less metric used in text-to-text tasks, that compares the generated summaries directly to the source text, by automatically asking and answering questions.
QuestEval: Summarization Asks for Fact-based Evaluation
Summarization evaluation remains an open research problem: current metrics such as ROUGE are known to be limited and to correlate poorly with human judgments.
Controlling Hallucinations at Word Level in Data-to-Text Generation
Specifically, we propose a Multi-Branch Decoder which is able to leverage word-level labels to learn the relevant parts of each training instance.
Ranked #3 on
Table-to-Text Generation
on WikiBio
Normalizing Kalman Filters for Multivariate Time Series Analysis
This paper tackles the modelling of large, complex and multivariate time series panels in a probabilistic setting.
Stochastic sparse adversarial attacks
This paper introduces stochastic sparse adversarial attacks (SSAA), standing as simple, fast and purely noise-based targeted and untargeted attacks of neural network classifiers (NNC).
Transductive Zero-Shot Learning using Cross-Modal CycleGAN
We show the efficiency of our Cross-Modal CycleGAN model (CM-GAN) on the ImageNet T-ZSL task where we obtain state-of-the-art results.
PARENTing via Model-Agnostic Reinforcement Learning to Correct Pathological Behaviors in Data-to-Text Generation
Evaluations on the widely used WikiBIO and WebNLG benchmarks demonstrate the effectiveness of this framework compared to state-of-the-art models.
Augmenting Physical Models with Deep Networks for Complex Dynamics Forecasting
In this work, we introduce the APHYNITY framework, a principled approach for augmenting incomplete physical dynamics described by differential equations with deep data-driven models.
Let's Stop Incorrect Comparisons in End-to-end Relation Extraction!
Despite efforts to distinguish three different evaluation setups (Bekoulis et al., 2018), numerous end-to-end Relation Extraction (RE) articles present unreliable performance comparison to previous work.
A Principle of Least Action for the Training of Neural Networks
From this observation, we reformulate the learning problem as follows: finding neural networks which solve the task while transporting the data as efficiently as possible.
PDE-Driven Spatiotemporal Disentanglement
A recent line of work in the machine learning community addresses the problem of predicting high-dimensional spatiotemporal phenomena by leveraging specific tools from the differential equations theory.
What BERT Sees: Cross-Modal Transfer for Visual Question Generation
Pre-trained language models have recently contributed to significant advances in NLP tasks.
Stochastic Latent Residual Video Prediction
Designing video prediction models that account for the inherent uncertainty of the future is challenging.
Ranked #1 on
Video Prediction
on Cityscapes 128x128
(Pred metric)
Incorporating Visual Semantics into Sentence Representations within a Grounded Space
To overcome this limitation, we propose to transfer visual information to textual representations by learning an intermediate representation space: the grounded space.
Contextualized Embeddings in Named-Entity Recognition: An Empirical Study on Generalization
Contextualized embeddings use unsupervised language model pretraining to compute word representations depending on their context.
A Hierarchical Model for Data-to-Text Generation
This however loses most of the structure contained in the data.
Unsupervised Adversarial Image Inpainting
This allows us sampling from the latent component in order to generate a distribution of images associated to an observation.
Unsupervised domain adaptation with imputation
Motivated by practical applications, we consider unsupervised domain adaptation for classification problems, in the presence of missing data in the target domain.
Unsupervised Spatiotemporal Data Inpainting
We tackle the problem of inpainting occluded area in spatiotemporal sequences, such as cloud occluded satellite observations, in an unsupervised manner.
Optimal Unsupervised Domain Translation
Domain Translation is the problem of finding a meaningful correspondence between two domains.
Benchmarking Regression Methods: A comparison with CGAN
Such a real world situation is best represented using an implicit model in which an extra noise vector, $z$ is included with $x$ as input.
Unsupervised Adversarial Image Reconstruction
We address the problem of recovering an underlying signal from lossy, inaccurate observations in an unsupervised setting.
Learning Partially Observed PDE Dynamics with Neural Networks
Spatio-Temporal processes bear a central importance in many applied scientific fields.
Copy mechanism and tailored training for character-based data-to-text generation
In the last few years, many different methods have been focusing on using deep recurrent neural networks for natural language generation.
Ranked #2 on
Data-to-Text Generation
on E2E NLG Challenge
Context-Aware Zero-Shot Learning for Object Recognition
Zero-Shot Learning (ZSL) aims at classifying unlabeled objects by leveraging auxiliary knowledge, such as semantic representations.
Learning Dynamical Systems from Partial Observations
We consider the problem of forecasting complex, nonlinear space-time processes when observations provide only partial information of on the system's state.
DEFT 2018: Attention s\'elective pour classification de microblogs (DEFT 2018 : Selective Attention for Microblogging Classification )
Dans le cadre de l{'}atelier DEFT 2018 nous nous sommes int{\'e}ress{\'e}s {\`a} la classification de microblogs (ici, des tweets) r{\'e}dig{\'e}s en fran{\c{c}}ais.
Spatio-Temporal Neural Networks for Space-Time Series Forecasting and Relations Discovery
We introduce a dynamical spatio-temporal model formalized as a recurrent neural network for forecasting time series of spatial processes, i. e. series of observations sharing temporal and spatial dependencies.
A Distributed Frank-Wolfe Framework for Learning Low-Rank Matrices with the Trace Norm
We consider the problem of learning a high-dimensional but low-rank matrix from a large-scale dataset distributed over several machines, where low-rankness is enforced by a convex trace norm constraint.
Deep Learning for Physical Processes: Incorporating Prior Scientific Knowledge
We consider the use of Deep Learning methods for modeling complex phenomena like those occurring in natural physical processes.
Learning Multi-Modal Word Representation Grounded in Visual Context
Representing the semantics of words is a long-standing problem for the natural language processing community.
OCReP: An Optimally Conditioned Regularization for Pseudoinversion Based Neural Training
In this paper we consider the training of single hidden layer neural networks by pseudoinversion, which, in spite of its popularity, is sometimes affected by numerical instability issues.
Extended Recommendation Framework: Generating the Text of a User Review as a Personalized Summary
We propose to augment rating based recommender systems by providing the user with additional information which might help him in his choice or in the understanding of the recommendation.
Deep Sequential Neural Network
Instead of considering global transformations, like in classical multilayer networks, this model allows us for learning a set of local transformations.
Learning States Representations in POMDP
We propose to deal with sequential processes where only partial observations are available by learning a latent representation space on which policies may be accurately learned.
Learning Information Spread in Content Networks
We introduce a model for predicting the diffusion of content information on social media.
Sequentially Generated Instance-Dependent Image Representations for Classification
In this paper, we investigate a new framework for image classification that adaptively generates spatial representations.
Robust Bloom Filters for Large MultiLabel Classification Tasks
This paper presents an approach to multilabel classification (MLC) with a large number of labels.
