Alexandre Drouin
Head of Frontier AI Research
AI Research Leadership
Alexandre Drouin is the Head of Frontier AI Research at ServiceNow AI Research and an Adjunct Professor of Computer Science at Laval University and Mila. He leads a team exploring the capabilities needed for reliable enterprise AI agents, including computer-use automation, data analytics, and decision-making, as well as the barriers to their adoption, such as security, trustworthiness, and rigorous evaluation. His research spans causal inference, probabilistic forecasting, and LLM-based agents, with recent contributions including benchmarks and frameworks for browser automation, deep research, and agent robustness. Beyond his work at ServiceNow, Alexandre is actively involved in the scientific community, notably as Program Chair for the NeurIPS 2026 Evaluations and Datasets Track. He holds a Ph.D. in Computer Science from Université Laval, supervised by François Laviolette, and joined ServiceNow AI Research from Element AI.
Interests
- Causality
- Time Series Forecasting
- Acting under Uncertainty
Publications
Causal Discovery With Metadata-Informed Latent Types.
Conference on Uncertainty in Artificial Intelligence (UAI),
2026.
Dr-CiK: A Testbed for Foresight-Driven Agents.
Workshop at the International Conference of Machine Learning (ICML),
2026.
WebArena-Pro: A Heterogeneous, Multimodal, Reproducible Benchmark for Web Agents.
Workshop at the International Conference of Machine Learning (ICML),
2026.
Overcoming the Modality Gap in Context-Aided Forecasting.
International Conference on Machine Learning (ICML),
2026.
Malice in Agentland: Down the Rabbit Hole of Backdoors in the AI Supply Chain.
ACM Conference on AI and Agentic Systems,
2026.
Overcoming the Modality Gap in Context-Aided Forecasting.
Workshop at the International Conference of Machine Learning (ICML),
2026.
DRBench: A Realistic Benchmark for Enterprise Deep Research.
International Conference on Learning Representations,
2026.
Learning a Spatial Partitioning and its Causal Relations from Temporal Data.
Causal Learning and Reasoning (CLeaR),
2026.
Beyond Naïve Prompting: Strategies for Improved Zero-shot Context-aided Forecasting with LLMs.
Workshop at the Neural Information Processing Systems (NeurIPS),
2025.
How to Train Your LLM Web Agent: A Statistical Diagnosis.
Workshop at the Neural Information Processing Systems (NeurIPS),
2025.
Causal Differentiating Concepts: Interpreting LM Behavior via Causal Representation Learning.
Neural Information Processing Systems (NeurIPS),
2025.
How to Train Your LLM Web Agent: A Statistical Diagnosis.
Neural Information Processing Systems (NeurIPS),
2025.
Beyond Naïve Prompting: Strategies for Improved Zero-shot Context-aided Forecasting with LLMs.
Conference on Language Modeling Workshops,
2025.
DoomArena: A framework for Testing AI Agents Against Evolving Security Threats.
Conference on Language Modeling (COLM),
2025.
DoomArena: A framework for Testing AI Agents Against Evolving Security Threats.
Workshop at the International Conference of Machine Learning (ICML),
2025.
How to Train Your LLM Web Agent: A Statistical Diagnosis (Oral).
Workshop at the International Conference of Machine Learning (ICML),
2025.
Silent Sabotage: Injecting Backdoors into AI Agents Through Fine-Tuning.
Workshop at the International Conference of Machine Learning (ICML),
2025.
Context is Key: A Benchmark for Forecasting with Essential Textual Information.
International Conference on Machine Learning (ICML),
2025.
Generalization Bounds via Meta-Learned Model Representations: PAC-Bayes and Sample Compression Hypernetworks.
International Conference on Machine Learning (ICML),
2025.
The Landscape of Causal Discovery Data: Grounding Causal Discovery in Real-World Applications.
Causal Learning and Reasoning (CLeaR),
2025.
Learning to Defer for Causal Discovery with Imperfect Experts.
Workshop at the International Conference of Learning Representation (ICLR),
2025.
The Landscape of Causal Discovery Data: Grounding Causal Discovery in Real-World Applications.
Workshop at the International Conference of Learning Representation (ICLR),
2025.
InsightBench: Evaluating Business Analytics Agents Through Multi-Step Insight Generation.
International Conference of Learning Representations (ICLR),
2025.
The BrowserGym Ecosystem for Web Agent Research.
Transactions on Machine Learning Research (TMLR),
2025.
Context is Key: A Benchmark for Forecasting with Essential Textual Information.
Workshop at the Neural Information Processing Systems (NeurIPS),
2024.
Evaluating Interventional Reasoning Capabilities of Large Language Models.
Workshop at the Neural Information Processing Systems (NeurIPS),
2024.
Fine-Tuning Web Agents: It Works, But It's Trickier Than You Think.
Workshop at the Neural Information Processing Systems (NeurIPS),
2024.
Sample Compression Hypernetworks: From Generalization Bounds to Meta-Learning.
Workshop at the Neural Information Processing Systems (NeurIPS),
2024.
WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks.
NeurIPS Datasets and Benchmarks Track (NeurIPS Datasets),
2024.
Context is Key: A Benchmark for Forecasting with Essential Textual Information.
Foundation Models for Time Series,
2024.
An Ecosystem for Web Agents: WorkArena, BrowserGym, AgentLab and more.
Montreal AI Symposium (MAIS),
2024.
Context is Key: A Benchmark for Forecasting with Essential Textual Information.
Montreal AI Symposium (MAIS),
2024.
TACTIS-2: Better, Faster, Simpler Attentional Copulas for Multivariate Time Series.
Montreal AI Symposium (MAIS),
2024.
WorkArena: How Capable are Web Agents at Solving Common Knowledge Work Tasks?.
International Conference on Machine Learning (ICML),
2024.
TACTIS-2: Better, Faster, Simpler Attentional Copulas for Multivariate Time Series.
International Conference of Learning Representations (ICLR),
2024.
WorkArena: How Capable are Web Agents at Solving Common Knowledge Work Tasks?.
Workshop at the International Conference of Learning Representation (ICLR),
2024.
Towards Disentangled High-level Causal Explanations in Text.
Mid-Atlantic Student Colloquium on Speech, Language and Learning,
2024.
Capture the Flag: Uncovering Data Insights with Large Language Models.
Workshop at the Neural Information Processing Systems (NeurIPS),
2023.
Lag-Llama: A Foundation Model for Probabilistic Time Series Forecasting.
Workshop at the Neural Information Processing Systems (NeurIPS),
2023.
The Unsolved Challenges of LLMs in Open-Ended Web Tasks: A Case Study.
Workshop at the Neural Information Processing Systems (NeurIPS),
2023.
GEO-Bench: Toward Foundation Models for Earth Monitoring.
NeurIPS Datasets and Benchmarks Track (NeurIPS Datasets),
2023.
Invariant Causal Set Covering Machines.
Workshop on Spurious Correlations, Invariance, and Stability (ICML),
2023.
Benchmarking Bayesian Causal Discovery Methods for Downstream Treatment Effect Estimation.
Workshop on Structured Probabilistic Inference & Generative Modeling (ICML),
2023.
Causal Discovery with Language Models as Imperfect Experts.
Workshop on Structured Probabilistic Inference & Generative Modeling (ICML),
2023.
Regions of Reliability in the Evaluation of Multivariate Probabilistic Forecasts.
International Conference on Machine Learning (ICML),
2023.
TACTiS: Transformer-Attentional Copulas for Time Series.
Montreal AI Symposium (MAIS),
2022.
TACTiS: Transformer-Attentional Copulas for Time Series.
International Conference on Machine Learning (ICML),
2022.
Deconfounding Dynamic Treatment Regimes.
SSC,
2022.
Typing assumptions improve identification in causal discovery - theory and algorithms.
Causal Learning and Reasoning (CLeaR),
2022.
Toward Foundation Models for Earth Monitoring: Proposal for a Climate Change Benchmark.
Workshop at the Neural Information Processing Systems (NeurIPS),
2021.
Typing assumptions improve identification in causal discovery - Report and comments on future directions.
Workshop at the Neural Information Processing Systems (NeurIPS),
2021.
RandomSCM: interpretable ensembles of sparse classifiers tailored for omics data.
Intelligent Systems for Molecular Biology and European Conference on Computational Biology,
2021.
Typing assumptions improve identification in causal discovery.
Workshop at the International Conference on Machine Learning (ICML),
2021.
Phylogenetic Manifold Regularization: a semi-supervised approach to predict transcription factor binding sites.
International Conference on Bioinformatics and Biomedicine (BIBM),
2020.
Differentiable Causal Discovery from Interventional Data.
Conference on Neural Information Processing Systems (NeurIPS),
2020.
In search of robust measures of generalization.
Conference on Neural Information Processing Systems (NeurIPS),
2020.
Synbols: Probing Learning Algorithms with Synthetic Datasets.
Conference on Neural Information Processing Systems (NeurIPS),
2020.
Embedding Propagation: Smoother Manifold for Few-Shot Classification.
European Conference on Computer Vision (ECCV),
2020.
Gradient-Based Neural DAG Learning with Interventions.
Workshop at the International Conference on Learning Representations (ICLR),
2020.
Mass spectra alignment using virtual lock-masses.
Nature Scientific Reports,
2019.
Interpretable genotype-to-phenotype classifiers with performance guarantees.
Nature Scientific Reports,
2018.
Synbols: Probing Learning Algorithms with Synthetic Datasets.
Montreal AI Symposium (MAIS),
2018.
Deep learning for electromyographic hand gesture signal classification using transfer learning.
Transactions on Neural Systems and Rehabilitation Engineering (IEEE NSRE),
2018.