arXiv subjects

Laure Berti-Equille

Publications and source records attributed to Laure Berti-Equille.

18 recordsLinked to original sources

arXiv2026

Model-Aware Data Cleaning for Tabular Foundation Models

Tabular Foundation Models (TFMs) achieve state-of-the-art zero-shot accuracy on small tabular datasets, but their in-context learning assumes approximately clean inputs: real- world missing values, outliers, and duplicates create a prior mismatch that degrades both accuracy and calibration. We study reinforcement learning for tabular data cleaning, a learned policy that sequences cleaning operators and introduce L2C-TFM with a model-aware reward (TFMAwareReward). We are explicit about what this reward optimizes: it regularizes the Wasserstein distance between the cleaned and the original (dirty) data, a distributional-stability term, which we measure as a diagnostic. Across six experiments on ten OpenML datasets: (i) three of seven reward designs collapse to degenerate strategies, so reward engineering is non-trivial; (ii) under an 8-seed repeated-holdout protocol the model-aware reward matches a random-forest-reward baseline on accuracy (p=0.38), with a benefit confined to minority-class macro-F1 under class imbalance that is partly a reward-agnostic calibrated-threshold effect; and (iii) a policy pre-trained on one dataset transfers to held-out datasets. A diagnostic analysis shows that the two distances are distinct objectives: cleaning tends to move data away from the prior, and prior-distance, not distance-to-dirty, is what tracks downstream quality. We therefore treat prior alignment as a motivating objective and a target for future work, not a property of the reward evaluated here. Code, datasets, and the nested evaluation harness are available at https://github.com/LaureBerti/Learn2Clean/tree/master/Learn2Clean_TFM.

cs.LG↗

arXiv2026

Data Quality Profiling at Scale with Progressive Sampling: A Benchmark for Data-Centric AI Pipelines

Data quality profiling -- computing missing-value rates, duplicate fractions, outlier densities, and functional-dependency violations -- is foundational for data-centric AI pipelines, yet exhaustive scans over millions of rows are prohibitively slow for near-real-time monitoring. Progressive sampling is the standard alternative; the open question is which strategy best preserves profile fidelity at scale. We benchmark nine sampling strategies -- blind (random uniform, geometric, Yamane, cluster) and proxy-guided (Metropolis-Hastings, DAG, stratified by column type or quality score, importance-weighted) -- on three real-world datasets (NYC 311, NYPD arrests, UCI Adult; up to 500K rows), an IoT sensor stream (2.3M rows), two ultra-large real datasets including Ultra-Marathon Running (up to 7.4M rows), and synthetic data scaled to 5x10^6 rows. Contrary to the assumption sharpens estimates, blind representative samplers dominate uniformly. At a 5% budget, random uniform achieves 0.49% mean relative error on NYC 311; DAG-guided MCMC yields 19.5% (approx. 40x worse), and across all real datasets DAG is 11-49x worse (Wilcoxon W=0, p=0.002, n=9 pairs). Cluster sampling matches random uniform (MRE 0.110 vs. 0.111); proxy-guided methods share DAG's failure mode (MRE 0.20-0.35). At scale, random uniform is near-linear (O(N^{0.964})) while DAG is super-linear (O(N^{1.272})), running 28--47x slower on ultra-large data with 6x worse accuracy. The root cause is an IQR proxy mismatch: proxy-guided samplers over-pursue numeric outliers, while quality defects concentrate in categorical columns invisible to the proxy. The actionable finding: representativeness, not domain knowledge, determines sampler quality -- schema-free random uniform or cluster sampling suffices for production-grade quality profiling at scale.

cs.DB↗

arXiv2025

Single Word Change is All You Need: Using LLMs to Create Synthetic Training Examples for Text Classifiers

In text classification, creating an adversarial example means subtly perturbing a few words in a sentence without changing its meaning, causing it to be misclassified by a classifier. A concerning observation is that a significant portion of adversarial examples generated by existing methods change only one word. This single-word perturbation vulnerability represents a significant weakness in classifiers, which malicious users can exploit to efficiently create a multitude of adversarial examples. This paper studies this problem and makes the following key contributions: (1) We introduce a novel metric $ρ$ to quantitatively assess a classifier's robustness against single-word perturbation. (2) We present the SP-Attack, designed to exploit the single-word perturbation vulnerability, achieving a higher attack success rate, better preserving sentence meaning, while reducing computation costs compared to state-of-the-art adversarial methods. (3) We propose SP-Defense, which aims to improve \r{ho} by applying data augmentation in learning. Experimental results on 4 datasets and BERT and distilBERT classifiers show that SP-Defense improves $ρ$ by 14.6% and 13.9% and decreases the attack success rate of SP-Attack by 30.4% and 21.2% on two classifiers respectively, and decreases the attack success rate of existing attack methods that involve multiple-word perturbations.

cs.CL↗

arXiv2024

Explingo: Explaining AI Predictions using Large Language Models

Explanations of machine learning (ML) model predictions generated by Explainable AI (XAI) techniques such as SHAP are essential for people using ML outputs for decision-making. We explore the potential of Large Language Models (LLMs) to transform these explanations into human-readable, narrative formats that align with natural communication. We address two key research questions: (1) Can LLMs reliably transform traditional explanations into high-quality narratives? and (2) How can we effectively evaluate the quality of narrative explanations? To answer these questions, we introduce Explingo, which consists of two LLM-based subsystems, a Narrator and Grader. The Narrator takes in ML explanations and transforms them into natural-language descriptions. The Grader scores these narratives on a set of metrics including accuracy, completeness, fluency, and conciseness. Our experiments demonstrate that LLMs can generate high-quality narratives that achieve high scores across all metrics, particularly when guided by a small number of human-labeled and bootstrapped examples. We also identified areas that remain challenging, in particular for effectively scoring narratives in complex domains. The findings from this work have been integrated into an open-source tool that makes narrative explanations available for further applications.

cs.CL↗

arXiv2024

OrionBench: Benchmarking Time Series Generative Models in the Service of the End-User

Time series anomaly detection is a vital task in many domains, including patient monitoring in healthcare, forecasting in finance, and predictive maintenance in energy industries. This has led to a proliferation of anomaly detection methods, including deep learning-based methods. Benchmarks are essential for comparing the performances of these models as they emerge, in a fair, rigorous, and reproducible approach. Although several benchmarks for comparing models have been proposed, these usually rely on a one-time execution over a limited set of datasets, with comparisons restricted to a few models. We propose OrionBench: an end-user centric, continuously maintained benchmarking framework for unsupervised time series anomaly detection models. Our framework provides universal abstractions to represent models, hyperparameter standardization, extensibility to add new pipelines and datasets, pipeline verification, and frequent releases with published updates of the benchmark. We demonstrate how to use OrionBench, and the performance of pipelines across 17 releases published over the course of four years. We also walk through two real scenarios we experienced with OrionBench that highlight the importance of continuous benchmarking for unsupervised time series anomaly detection.

cs.LG↗

arXiv2024

Large language models can be zero-shot anomaly detectors for time series?

Recent studies have shown the ability of large language models to perform a variety of tasks, including time series forecasting. The flexible nature of these models allows them to be used for many applications. In this paper, we present a novel study of large language models used for the challenging task of time series anomaly detection. This problem entails two aspects novel for LLMs: the need for the model to identify part of the input sequence (or multiple parts) as anomalous; and the need for it to work with time series data rather than the traditional text input. We introduce sigllm, a framework for time series anomaly detection using large language models. Our framework includes a time-series-to-text conversion module, as well as end-to-end pipelines that prompt language models to perform time series anomaly detection. We investigate two paradigms for testing the abilities of large language models to perform the detection task. First, we present a prompt-based detection method that directly asks a language model to indicate which elements of the input are anomalies. Second, we leverage the forecasting capability of a large language model to guide the anomaly detection process. We evaluated our framework on 11 datasets spanning various sources and 10 pipelines. We show that the forecasting method significantly outperformed the prompting method in all 11 datasets with respect to the F1 score. Moreover, while large language models are capable of finding anomalies, state-of-the-art deep learning models are still superior in performance, achieving results 30% better than large language models.

cs.LG↗

arXiv2023

Pyreal: A Framework for Interpretable ML Explanations

Users in many domains use machine learning (ML) predictions to help them make decisions. Effective ML-based decision-making often requires explanations of ML models and their predictions. While there are many algorithms that explain models, generating explanations in a format that is comprehensible and useful to decision-makers is a nontrivial task that can require extensive development overhead. We developed Pyreal, a highly extensible system with a corresponding Python implementation for generating a variety of interpretable ML explanations. Pyreal converts data and explanations between the feature spaces expected by the model, relevant explanation algorithms, and human users, allowing users to generate interpretable explanations in a low-code manner. Our studies demonstrate that Pyreal generates more useful explanations than existing systems while remaining both easy-to-use and efficient.

cs.LG↗

arXiv2023

Discovering Transition Pathways Towards Coviability with Machine Learning

Coviability refers to the multiple socio-ecological arrangements and governance structures under which humans and nature can coexist in functional, fair, and persistent ways. Transitioning to a coviable state in environmentally degraded and socially vulnerable territories is challenging. This paper presents an ongoing French-Brazilian joint research project combining machine learning, agroecology, and social sciences to discover coviability pathways that can be adopted and implemented by local populations in the North-East region of Brazil.

cs.CY↗

arXiv2022

AER: Auto-Encoder with Regression for Time Series Anomaly Detection

Anomaly detection on time series data is increasingly common across various industrial domains that monitor metrics in order to prevent potential accidents and economic losses. However, a scarcity of labeled data and ambiguous definitions of anomalies can complicate these efforts. Recent unsupervised machine learning methods have made remarkable progress in tackling this problem using either single-timestamp predictions or time series reconstructions. While traditionally considered separately, these methods are not mutually exclusive and can offer complementary perspectives on anomaly detection. This paper first highlights the successes and limitations of prediction-based and reconstruction-based methods with visualized time series signals and anomaly scores. We then propose AER (Auto-encoder with Regression), a joint model that combines a vanilla auto-encoder and an LSTM regressor to incorporate the successes and address the limitations of each method. Our model can produce bi-directional predictions while simultaneously reconstructing the original time series by optimizing a joint objective function. Furthermore, we propose several ways of combining the prediction and reconstruction errors through a series of ablation studies. Finally, we compare the performance of the AER architecture against two prediction-based methods and three reconstruction-based methods on 12 well-known univariate time series datasets from NASA, Yahoo, Numenta, and UCR. The results show that AER has the highest averaged F1 score across all datasets (a 23.5% improvement compared to ARIMA) while retaining a runtime similar to its vanilla auto-encoder and regressor components. Our model is available in Orion, an open-source benchmarking tool for time series anomaly detection.

cs.LG↗

arXiv2022

R&R: Metric-guided Adversarial Sentence Generation

Adversarial examples are helpful for analyzing and improving the robustness of text classifiers. Generating high-quality adversarial examples is a challenging task as it requires generating fluent adversarial sentences that are semantically similar to the original sentences and preserve the original labels, while causing the classifier to misclassify them. Existing methods prioritize misclassification by maximizing each perturbation's effectiveness at misleading a text classifier; thus, the generated adversarial examples fall short in terms of fluency and similarity. In this paper, we propose a rewrite and rollback (R&R) framework for adversarial attack. It improves the quality of adversarial examples by optimizing a critique score which combines the fluency, similarity, and misclassification metrics. R&R generates high-quality adversarial examples by allowing exploration of perturbations that do not have immediate impact on the misclassification metric but can improve fluency and similarity metrics. We evaluate our method on 5 representative datasets and 3 classifier architectures. Our method outperforms current state-of-the-art in attack success rate by +16.2%, +12.8%, and +14.0% on the classifiers respectively. Code is available at https://github.com/DAI-Lab/fibber

cs.CL↗

arXiv2022

Reconstruction of Long-Term Historical Demand Data

Long-term planning of a robust power system requires the understanding of changing demand patterns. Electricity demand is highly weather sensitive. Thus, the supply side variation from introducing intermittent renewable sources, juxtaposed with variable demand, will introduce additional challenges in the grid planning process. By understanding the spatial and temporal variability of temperature over the US, the response of demand to natural variability and climate change-related effects on temperature can be separated, especially because the effects due to the former factor are not known. Through this project, we aim to better support the technology & policy development process for power systems by developing machine and deep learning 'back-forecasting' models to reconstruct multidecadal demand records and study the natural variability of temperature and its influence on demand.

cs.LG↗

arXiv2022

Sintel: A Machine Learning Framework to Extract Insights from Signals

The detection of anomalies in time series data is a critical task with many monitoring applications. Existing systems often fail to encompass an end-to-end detection process, to facilitate comparative analysis of various anomaly detection methods, or to incorporate human knowledge to refine output. This precludes current methods from being used in real-world settings by practitioners who are not ML experts. In this paper, we introduce Sintel, a machine learning framework for end-to-end time series tasks such as anomaly detection. The framework uses state-of-the-art approaches to support all steps of the anomaly detection process. Sintel logs the entire anomaly detection journey, providing detailed documentation of anomalies over time. It enables users to analyze signals, compare methods, and investigate anomalies through an interactive visualization tool, where they can annotate, modify, create, and remove events. Using these annotations, the framework leverages human knowledge to improve the anomaly detection pipeline. We demonstrate the usability, efficiency, and effectiveness of Sintel through a series of experiments on three public time series datasets, as well as one real-world use case involving spacecraft experts tasked with anomaly analysis tasks. Sintel's framework, code, and datasets are open-sourced at https://github.com/sintel-dev/.

cs.LG↗

arXiv2022

The Need for Interpretable Features: Motivation and Taxonomy

Through extensive experience developing and explaining machine learning (ML) applications for real-world domains, we have learned that ML models are only as interpretable as their features. Even simple, highly interpretable model types such as regression models can be difficult or impossible to understand if they use uninterpretable features. Different users, especially those using ML models for decision-making in their domains, may require different levels and types of feature interpretability. Furthermore, based on our experiences, we claim that the term "interpretable feature" is not specific nor detailed enough to capture the full extent to which features impact the usefulness of ML explanations. In this paper, we motivate and discuss three key lessons: 1) more attention should be given to what we refer to as the interpretable feature space, or the state of features that are useful to domain experts taking real-world actions, 2) a formal taxonomy is needed of the feature properties that may be required by these domain experts (we propose a partial taxonomy in this paper), and 3) transforms that take data from the model-ready state to an interpretable form are just as essential as traditional ML transforms that prepare features for the model.

cs.LG↗

arXiv2019

Are Outlier Detection Methods Resilient to Sampling?

Outlier detection is a fundamental task in data mining and has many applications including detecting errors in databases. While there has been extensive prior work on methods for outlier detection, modern datasets often have sizes that are beyond the ability of commonly used methods to process the data within a reasonable time. To overcome this issue, outlier detection methods can be trained over samples of the full-sized dataset. However, it is not clear how a model trained on a sample compares with one trained on the entire dataset. In this paper, we introduce the notion of resilience to sampling for outlier detection methods. Orthogonal to traditional performance metrics such as precision/recall, resilience represents the extent to which the outliers detected by a method applied to samples from a sampling scheme matches those when applied to the whole dataset. We propose a novel approach for estimating the resilience to sampling of both individual outlier methods and their ensembles. We performed an extensive experimental study on synthetic and real-world datasets where we study seven diverse and representative outlier detection methods, compare results obtained from samples versus those obtained from the whole datasets and evaluate the accuracy of our resilience estimates. We observed that the methods are not equally resilient to a given sampling scheme and it is often the case that careful joint selection of both the sampling scheme and the outlier detection method is necessary. It is our hope that the paper initiates research on designing outlier detection algorithms that are resilient to sampling.

cs.LG↗

arXiv2016

Using Twitter to Understand Public Interest in Climate Change: The case of Qatar

Climate change has received an extensive attention from public opinion in the last couple of years, after being considered for decades as an exclusive scientific debate. Governments and world-wide organizations such as the United Nations are working more than ever on raising and maintaining public awareness toward this global issue. In the present study, we examine and analyze Climate Change conversations in Qatar's Twittersphere, and sense public awareness towards this global and shared problem in general, and its various related topics in particular. Such topics include but are not limited to politics, economy, disasters, energy and sandstorms. To address this concern, we collect and analyze a large dataset of 109 million tweets posted by 98K distinct users living in Qatar -- one of the largest emitters of CO2 worldwide. We use a taxonomy of climate change topics created as part of the United Nations Pulse project to capture the climate change discourse in more than 36K tweets. We also examine which topics people refer to when they discuss climate change, and perform different analysis to understand the temporal dynamics of public interest toward these topics.

cs.SI↗

arXiv2015

Data Fusion: Resolving Conflicts from Multiple Sources

Many data management applications, such as setting up Web portals, managing enterprise data, managing community data, and sharing scientific data, require integrating data from multiple sources. Each of these sources provides a set of values and different sources can often provide conflicting values. To present quality data to users, it is critical to resolve conflicts and discover values that reflect the real world; this task is called {\em data fusion}. This paper describes a novel approach that finds true values from conflicting information when there are a large number of sources, among which some may copy from others. We present a case study on real-world data showing that the described algorithm can significantly improve accuracy of truth discovery and is scalable when there are a large number of data sources.

cs.DB↗

arXiv2014

Truth Discovery Algorithms: An Experimental Evaluation

A fundamental problem in data fusion is to determine the veracity of multi-source data in order to resolve conflicts. While previous work in truth discovery has proved to be useful in practice for specific settings, sources' behavior or data set characteristics, there has been limited systematic comparison of the competing methods in terms of efficiency, usability, and repeatability. We remedy this deficit by providing a comprehensive review of 12 state-of-the art algorithms for truth discovery. We provide reference implementations and an in-depth evaluation of the methods based on extensive experiments on synthetic and real-world data. We analyze aspects of the problem that have not been explicitly studied before, such as the impact of initialization and parameter setting, convergence, and scalability. We provide an experimental framework for extensively comparing the methods in a wide range of truth discovery scenarios where source coverage, numbers and distributions of conflicts, and true positive claims can be controlled and used to evaluate the quality and performance of the algorithms. Finally, we report comprehensive findings obtained from the experiments and provide new insights for future research.

cs.DB↗

arXiv2009

Sailing the Information Ocean with Awareness of Currents: Discovery and Application of Source Dependence

The Web has enabled the availability of a huge amount of useful information, but has also eased the ability to spread false information and rumors across multiple sources, making it hard to distinguish between what is true and what is not. Recent examples include the premature Steve Jobs obituary, the second bankruptcy of United airlines, the creation of Black Holes by the operation of the Large Hadron Collider, etc. Since it is important to permit the expression of dissenting and conflicting opinions, it would be a fallacy to try to ensure that the Web provides only consistent information. However, to help in separating the wheat from the chaff, it is essential to be able to determine dependence between sources. Given the huge number of data sources and the vast volume of conflicting data available on the Web, doing so in a scalable manner is extremely challenging and has not been addressed by existing work yet. In this paper, we present a set of research problems and propose some preliminary solutions on the issues involved in discovering dependence between sources. We also discuss how this knowledge can benefit a variety of technologies, such as data integration and Web 2.0, that help users manage and access the totality of the available information from various sources.

cs.DB↗