Alexandre Drouin

Yihong Tang, Andrew Robert Williams, Arjun Ashok et al.

Time series forecasting in real-world settings often depends not only on historical observations, but also on external context that must be actively discovered from noisy, heterogeneous information sources. Yet existing context-aided forecasting benchmarks typically assume that the supporting context is already provided, leaving open whether agents can identify it on their own. Therefore, we introduce Dr-CiK, a benchmark for evaluating whether agents can retrieve forecasting-relevant supporting context from a document corpus, filter out distractors, distill the retrieved context into forecast-useful evidence, and generate forecasts supported by that evidence. Through context ablations and evaluations of state-of-the-art deep research and forecasting methods paired together, we show that high-quality context substantially improves forecasting performance in Dr-CiK. However, most existing DR agents recover only a small fraction of the ground-truth supporting evidence (usually <5%), are frequently misled by distractors (>80% distractor citations), and can cause forecasters to perform worse with retrieved context than without context. Our results motivate research on foresight-driven agents that search for the right context to predict the future.

Gaurav Sahu, Abhay Puri, Juan Rodriguez et al. · mila

Data analytics is essential for extracting valuable insights from data that can assist organizations in making effective decisions. We introduce InsightBench, a benchmark dataset with three key features. First, it consists of 100 datasets representing diverse business use cases such as finance and incident management, each accompanied by a carefully curated set of insights planted in the datasets. Second, unlike existing benchmarks focusing on answering single queries, InsightBench evaluates agents based on their ability to perform end-to-end data analytics, including formulating questions, interpreting answers, and generating a summary of insights and actionable steps. Third, we conducted comprehensive quality assurance to ensure that each dataset in the benchmark had clear goals and included relevant and meaningful questions and analysis. Furthermore, we implement a two-way evaluation mechanism using LLaMA-3 as an effective, open-source evaluator to assess agents' ability to extract insights. We also propose AgentPoirot, our baseline data analysis agent capable of performing end-to-end data analytics. Our evaluation on InsightBench shows that AgentPoirot outperforms existing approaches (such as Pandas Agent) that focus on resolving single queries. We also compare the performance of open- and closed-source LLMs and various evaluation strategies. Overall, this benchmark serves as a testbed to motivate further development in comprehensive automated data analytics and can be accessed here: https://github.com/ServiceNow/insight-bench.

Alexander Gurung, Spandana Gella, Alexandre Drouin et al.

Deep research agents increasingly combine private local documents with external tools like web retrieval, creating a privacy risk: an agent's external queries may leak sensitive information from its local context. This risk is amplified by the mosaic effect, where individual queries may appear harmless but become revealing in aggregate. We introduce MosaicLeaks, a benchmark of 1,001 multi-hop deep research tasks that chain private enterprise documents and a public web corpus, forcing agents to make external queries that depend on local information. We evaluate leakage with an adversary LLM that observes only the agent's external queries and attempts to infer private information at three levels: the agent's research intent, answers to specific private questions and verifiable claims about the enterprise documents. We find that models across families and sizes frequently leak at all three levels, that zero-shot privacy prompting reduces but does not eliminate leakage and that reinforcement learning for task performance alone worsens leakage. To address this, we propose Privacy-Aware Deep Research (PA-DR), an RL framework that combines situational rewards for task success with a learned privacy classifier to provide dense credit assignment over both per-query and mosaic-level leakage. Training Qwen3-4B-Instruct with PA-DR improves accuracy from 48.7% to 58.7% and reduces answer and full-information leakage from 34.0% to 9.9%.

Léo Boisvert, Megh Thakkar, Maxime Gasse et al.

The ability of large language models (LLMs) to mimic human-like intelligence has led to a surge in LLM-based autonomous agents. Though recent LLMs seem capable of planning and reasoning given user instructions, their effectiveness in applying these capabilities for autonomous task solving remains underexplored. This is especially true in enterprise settings, where automated agents hold the promise of a high impact. To fill this gap, we propose WorkArena++, a novel benchmark consisting of 682 tasks corresponding to realistic workflows routinely performed by knowledge workers. WorkArena++ is designed to evaluate the planning, problem-solving, logical/arithmetic reasoning, retrieval, and contextual understanding abilities of web agents. Our empirical studies across state-of-the-art LLMs and vision-language models (VLMs), as well as human workers, reveal several challenges for such models to serve as useful assistants in the workplace. In addition to the benchmark, we provide a mechanism to effortlessly generate thousands of ground-truth observation/action traces, which can be used for fine-tuning existing models. Overall, we expect this work to serve as a useful resource to help the community progress toward capable autonomous agents. The benchmark can be found at https://github.com/ServiceNow/WorkArena.

Alexandre Lacoste, Nils Lehmann, Pau Rodriguez et al.

Recent progress in self-supervision has shown that pre-training large neural networks on vast amounts of unsupervised data can lead to substantial increases in generalization to downstream tasks. Such models, recently coined foundation models, have been transformational to the field of natural language processing. Variants have also been proposed for image data, but their applicability to remote sensing tasks is limited. To stimulate the development of foundation models for Earth monitoring, we propose a benchmark comprised of six classification and six segmentation tasks, which were carefully curated and adapted to be both relevant to the field and well-suited for model evaluation. We accompany this benchmark with a robust methodology for evaluating models and reporting aggregated results to enable a reliable assessment of progress. Finally, we report results for 20 baselines to gain information about the performance of existing models. We believe that this benchmark will be a driver of progress across a variety of Earth monitoring tasks.

Arjun Ashok, Étienne Marcotte, Valentina Zantedeschi et al.

We introduce a new model for multivariate probabilistic time series prediction, designed to flexibly address a range of tasks including forecasting, interpolation, and their combinations. Building on copula theory, we propose a simplified objective for the recently-introduced transformer-based attentional copulas (TACTiS), wherein the number of distributional parameters now scales linearly with the number of variables instead of factorially. The new objective requires the introduction of a training curriculum, which goes hand-in-hand with necessary changes to the original architecture. We show that the resulting model has significantly better training dynamics and achieves state-of-the-art performance across diverse real-world forecasting tasks, while maintaining the flexibility of prior work, such as seamless handling of unaligned and unevenly-sampled time series. Code is made available at https://github.com/ServiceNow/TACTiS.

Kashif Rasul, Arjun Ashok, Andrew Robert Williams et al.

Over the past years, foundation models have caused a paradigm shift in machine learning due to their unprecedented capabilities for zero-shot and few-shot generalization. However, despite the success of foundation models in modalities such as natural language processing and computer vision, the development of foundation models for time series forecasting has lagged behind. We present Lag-Llama, a general-purpose foundation model for univariate probabilistic time series forecasting based on a decoder-only transformer architecture that uses lags as covariates. Lag-Llama is pretrained on a large corpus of diverse time series data from several domains, and demonstrates strong zero-shot generalization capabilities compared to a wide range of forecasting models on downstream datasets across domains. Moreover, when fine-tuned on relatively small fractions of such previously unseen datasets, Lag-Llama achieves state-of-the-art performance, outperforming prior deep learning approaches, emerging as the best general-purpose model on average. Lag-Llama serves as a strong contender to the current state-of-art in time series forecasting and paves the way for future advancements in foundation models tailored to time series data.

Chris Chinenye Emezue, Alexandre Drouin, Tristan Deleu et al.

The practical utility of causality in decision-making is widespread and brought about by the intertwining of causal discovery and causal inference. Nevertheless, a notable gap exists in the evaluation of causal discovery methods, where insufficient emphasis is placed on downstream inference. To address this gap, we evaluate seven established baseline causal discovery methods including a newly proposed method based on GFlowNets, on the downstream task of treatment effect estimation. Through the implementation of a distribution-level evaluation, we offer valuable and unique insights into the efficacy of these causal discovery methods for treatment effect estimation, considering both synthetic and real-world scenarios, as well as low-data scenarios. The results of our study demonstrate that some of the algorithms studied are able to effectively capture a wide range of useful and diverse ATE modes, while some tend to learn many low-probability modes which impacts the (unrelaxed) recall and precision.

Stephanie Long, Alexandre Piché, Valentina Zantedeschi et al.

Understanding the causal relationships that underlie a system is a fundamental prerequisite to accurate decision-making. In this work, we explore how expert knowledge can be used to improve the data-driven identification of causal graphs, beyond Markov equivalence classes. In doing so, we consider a setting where we can query an expert about the orientation of causal relationships between variables, but where the expert may provide erroneous information. We propose strategies for amending such expert knowledge based on consistency properties, e.g., acyclicity and conditional independencies in the equivalence class. We then report a case study, on real data, where a large language model is used as an imperfect expert.

Thibaud Godon, Pier-Luc Plante, Baptiste Bauvin et al.

Background: Understanding the relationship between the Omics and the phenotype is a central problem in precision medicine. The high dimensionality of metabolomics data challenges learning algorithms in terms of scalability and generalization. Most learning algorithms do not produce interpretable models -- Method: We propose an ensemble learning algorithm based on conjunctions or disjunctions of decision rules. -- Results : Applications on metabolomics data shows that it produces models that achieves high predictive performances. The interpretability of the models makes them useful for biomarker discovery and patterns discovery in high dimensional data.

Étienne Marcotte, Valentina Zantedeschi, Alexandre Drouin et al.

Multivariate probabilistic time series forecasts are commonly evaluated via proper scoring rules, i.e., functions that are minimal in expectation for the ground-truth distribution. However, this property is not sufficient to guarantee good discrimination in the non-asymptotic regime. In this paper, we provide the first systematic finite-sample study of proper scoring rules for time-series forecasting evaluation. Through a power analysis, we identify the "region of reliability" of a scoring rule, i.e., the set of practical conditions where it can be relied on to identify forecasting errors. We carry out our analysis on a comprehensive synthetic benchmark, specifically designed to test several key discrepancies between ground-truth and forecast distributions, and we gauge the generalizability of our findings to real-world tasks with an application to an electricity production problem. Our results reveal critical shortcomings in the evaluation of multivariate probabilistic forecasts as commonly performed in the literature.

Thibaud Godon, Baptiste Bauvin, Pascal Germain et al.

Rule-based models, such as decision trees, appeal to practitioners due to their interpretable nature. However, the learning algorithms that produce such models are often vulnerable to spurious associations and thus, they are not guaranteed to extract causally-relevant insights. In this work, we build on ideas from the invariant causal prediction literature to propose Invariant Causal Set Covering Machines, an extension of the classical Set Covering Machine algorithm for conjunctions/disjunctions of binary-valued rules that provably avoids spurious associations. We demonstrate both theoretically and empirically that our method can identify the causal parents of a variable of interest in polynomial time.

Alexandre Lacoste, Nicolas Gontier, Oleh Shliazhko et al. · ibm-research

The proliferation of agent benchmarks has created critical fragmentation that threatens research productivity. Each new benchmark requires substantial custom integration, creating an "integration tax" that limits comprehensive evaluation. We propose CUBE (Common Unified Benchmark Environments), a universal protocol standard built on MCP and Gym that allows benchmarks to be wrapped once and used everywhere. By separating task, benchmark, package, and registry concerns into distinct API layers, CUBE enables any compliant platform to access any compliant benchmark for evaluation, RL training, or data generation without custom integration. We call on the community to contribute to the development of this standard before platform-specific implementations deepen fragmentation as benchmark production accelerates through 2026.

Dheeraj Vattikonda, Santhoshi Ravichandran, Emiliano Penaloza et al. · mila

LLM-based web agents have recently made significant progress, but much of it has occurred in closed-source systems, widening the gap with open-source alternatives. Progress has been held back by two key challenges: first, a narrow focus on single-step tasks that overlooks the complexity of multi-step web interactions; and second, the high compute costs required to post-train LLM-based web agents. To address this, we present the first statistically grounded study on compute allocation for LLM web-agent post-training. Our approach uses a two-stage pipeline, training a Llama 3.1 8B student to imitate a Llama 3.3 70B teacher via supervised fine-tuning (SFT), followed by on-policy reinforcement learning. We find this process highly sensitive to hyperparameter choices, making exhaustive sweeps impractical. To spare others from expensive trial-and-error, we sample 1,370 configurations and use bootstrapping to estimate effective hyperparameters. Our results show that combining SFT with on-policy RL consistently outperforms either approach alone on both WorkArena and MiniWob++. Further, this strategy requires only 55% of the compute to match the peak performance of pure SFT on MiniWob++, effectively pushing the compute-performance Pareto frontier, and is the only strategy that can close the gap with closed-source models.

Amirhossein Abaskohi, Tianyi Chen, Miguel Muñoz-Mármol et al. · mila

We introduce DRBench, a benchmark for evaluating AI agents on complex, open-ended deep research tasks in enterprise settings. Unlike prior benchmarks that focus on simple questions or web-only queries, DRBench evaluates agents on multi-step queries (for example, ``What changes should we make to our product roadmap to ensure compliance with this standard?") that require identifying supporting facts from both the public web and private company knowledge base. Each task is grounded in realistic user personas and enterprise context, spanning a heterogeneous search space that includes productivity software, cloud file systems, emails, chat conversations, and the open web. Tasks are generated through a carefully designed synthesis pipeline with human-in-the-loop verification, and agents are evaluated on their ability to recall relevant insights, maintain factual accuracy, and produce coherent, well-structured reports. We release 15 deep research tasks across 10 domains, such as Sales, Cybersecurity, and Compliance. We demonstrate the effectiveness of DRBench by evaluating diverse DR agents across open- and closed-source models (such as GPT, Llama, and Qwen) and DR strategies, highlighting their strengths, weaknesses, and the critical path for advancing enterprise deep research. Code is available at https://github.com/ServiceNow/drbench.

Alexandre Drouin, Maxime Gasse, Massimo Caccia et al.

We study the use of large language model-based agents for interacting with software via web browsers. Unlike prior work, we focus on measuring the agents' ability to perform tasks that span the typical daily work of knowledge workers utilizing enterprise software systems. To this end, we propose WorkArena, a remote-hosted benchmark of 33 tasks based on the widely-used ServiceNow platform. We also introduce BrowserGym, an environment for the design and evaluation of such agents, offering a rich set of actions as well as multimodal observations. Our empirical evaluation reveals that while current agents show promise on WorkArena, there remains a considerable gap towards achieving full task automation. Notably, our analysis uncovers a significant performance disparity between open and closed-source LLMs, highlighting a critical area for future exploration and development in the field.

Hadi Nekoei, Aman Jaiswal, Patrice Bechard et al.

Large language model (LLM) agents perform well in sequential decision-making tasks, but improving them on unfamiliar domains often requires costly online interactions or fine-tuning on large expert datasets. These strategies are impractical for closed-source models and expensive for open-source ones, with risks of catastrophic forgetting. Offline trajectories offer reusable knowledge, yet demonstration-based methods struggle because raw traces are long, noisy, and tied to specific tasks. We present Just-in-time Episodic Feedback Hinter (JEF Hinter), an agentic system that distills offline traces into compact, context-aware hints. A zooming mechanism highlights decisive steps in long trajectories, capturing both strategies and pitfalls. Unlike prior methods, JEF Hinter leverages both successful and failed trajectories, extracting guidance even when only failure data is available, while supporting parallelized hint generation and benchmark-independent prompting. At inference, a retriever selects relevant hints for the current state, providing targeted guidance with transparency and traceability. Experiments on MiniWoB++, WorkArena-L1, and WebArena-Lite show that JEF Hinter consistently outperforms strong baselines, including human- and document-based hints.

Pau Rodríguez, Issam Laradji, Alexandre Drouin et al.

Few-shot classification is challenging because the data distribution of the training set can be widely different to the test set as their classes are disjoint. This distribution shift often results in poor generalization. Manifold smoothing has been shown to address the distribution shift problem by extending the decision boundaries and reducing the noise of the class representations. Moreover, manifold smoothness is a key factor for semi-supervised learning and transductive learning algorithms. In this work, we propose to use embedding propagation as an unsupervised non-parametric regularizer for manifold smoothing in few-shot classification. Embedding propagation leverages interpolations between the extracted features of a neural network based on a similarity graph. We empirically show that embedding propagation yields a smoother embedding manifold. We also show that applying embedding propagation to a transductive classifier achieves new state-of-the-art results in mini-Imagenet, tiered-Imagenet, Imagenet-FS, and CUB. Furthermore, we show that embedding propagation consistently improves the accuracy of the models in multiple semi-supervised learning scenarios by up to 16\% points. The proposed embedding propagation operation can be easily integrated as a non-parametric layer into a neural network. We provide the training code and usage examples at https://github.com/ElementAI/embedding-propagation.

Thibault Le Sellier De Chezelles, Maxime Gasse, Alexandre Drouin et al. · mila

The BrowserGym ecosystem addresses the growing need for efficient evaluation and benchmarking of web agents, particularly those leveraging automation and Large Language Models (LLMs). Many existing benchmarks suffer from fragmentation and inconsistent evaluation methodologies, making it challenging to achieve reliable comparisons and reproducible results. In an earlier work, Drouin et al. (2024) introduced BrowserGym which aims to solve this by providing a unified, gym-like environment with well-defined observation and action spaces, facilitating standardized evaluation across diverse benchmarks. We propose an extended BrowserGym-based ecosystem for web agent research, which unifies existing benchmarks from the literature and includes AgentLab, a complementary framework that aids in agent creation, testing, and analysis. Our proposed ecosystem offers flexibility for integrating new benchmarks while ensuring consistent evaluation and comprehensive experiment management. As a supporting evidence, we conduct the first large-scale, multi-benchmark web agent experiment and compare the performance of 6 state-of-the-art LLMs across 6 popular web agent benchmarks made available in BrowserGym. Among other findings, our results highlight a large discrepancy between OpenAI and Anthropic's latests models, with Claude-3.5-Sonnet leading the way on almost all benchmarks, except on vision-related tasks where GPT-4o is superior. Despite these advancements, our results emphasize that building robust and efficient web agents remains a significant challenge, due to the inherent complexity of real-world web environments and the limitations of current models.

Andrew Robert Williams, Arjun Ashok, Étienne Marcotte et al.

Forecasting is a critical task in decision-making across numerous domains. While historical numerical data provide a start, they fail to convey the complete context for reliable and accurate predictions. Human forecasters frequently rely on additional information, such as background knowledge and constraints, which can efficiently be communicated through natural language. However, in spite of recent progress with LLM-based forecasters, their ability to effectively integrate this textual information remains an open question. To address this, we introduce "Context is Key" (CiK), a time-series forecasting benchmark that pairs numerical data with diverse types of carefully crafted textual context, requiring models to integrate both modalities; crucially, every task in CiK requires understanding textual context to be solved successfully. We evaluate a range of approaches, including statistical models, time series foundation models, and LLM-based forecasters, and propose a simple yet effective LLM prompting method that outperforms all other tested methods on our benchmark. Our experiments highlight the importance of incorporating contextual information, demonstrate surprising performance when using LLM-based forecasting models, and also reveal some of their critical shortcomings. This benchmark aims to advance multimodal forecasting by promoting models that are both accurate and accessible to decision-makers with varied technical expertise. The benchmark can be visualized at https://servicenow.github.io/context-is-key-forecasting/v0/.

Philippe Brouillard, Chandler Squires, Jonas Wahl et al.

Causal discovery aims to automatically uncover causal relationships from data, a capability with significant potential across many scientific disciplines. However, its real-world applications remain limited. Current methods often rely on unrealistic assumptions and are evaluated only on simple synthetic toy datasets, often with inadequate evaluation metrics. In this paper, we substantiate these claims by performing a systematic review of the recent causal discovery literature. We present applications in biology, neuroscience, and Earth sciences - fields where causal discovery holds promise for addressing key challenges. We highlight available simulated and real-world datasets from these domains and discuss common assumption violations that have spurred the development of new methods. Our goal is to encourage the community to adopt better evaluation practices by utilizing realistic datasets and more adequate metrics.

Tejas Kasetty, Divyat Mahajan, Gintare Karolina Dziugaite et al.

Numerous decision-making tasks require estimating causal effects under interventions on different parts of a system. As practitioners consider using large language models (LLMs) to automate decisions, studying their causal reasoning capabilities becomes crucial. A recent line of work evaluates LLMs ability to retrieve commonsense causal facts, but these evaluations do not sufficiently assess how LLMs reason about interventions. Motivated by the role that interventions play in causal inference, in this paper, we conduct empirical analyses to evaluate whether LLMs can accurately update their knowledge of a data-generating process in response to an intervention. We create benchmarks that span diverse causal graphs (e.g., confounding, mediation) and variable types, and enable a study of intervention-based reasoning. These benchmarks allow us to isolate the ability of LLMs to accurately predict changes resulting from their ability to memorize facts or find other shortcuts. We evaluate six LLMs on the benchmarks, finding that GPT models show promising accuracy at predicting the intervention effects.

Issam Laradji, Perouz Taslakian, Sai Rajeswar et al.

The extraction of a small number of relevant insights from vast amounts of data is a crucial component of data-driven decision-making. However, accomplishing this task requires considerable technical skills, domain expertise, and human labor. This study explores the potential of using Large Language Models (LLMs) to automate the discovery of insights in data, leveraging recent advances in reasoning and code generation techniques. We propose a new evaluation methodology based on a "capture the flag" principle, measuring the ability of such models to recognize meaningful and pertinent information (flags) in a dataset. We further propose two proof-of-concept agents, with different inner workings, and compare their ability to capture such flags in a real-world sales dataset. While the work reported here is preliminary, our results are sufficiently interesting to mandate future exploration by the community.

Rafael Pardinas, Ehsan Kamalloo, David Vazquez et al.

Building general-purpose reasoning models using reinforcement learning with verifiable rewards (RLVR) across diverse domains has been widely adopted by frontier open-weight models. However, their training recipes and domain mixtures are often not disclosed. Joint optimization across domains poses significant challenges: domains vary widely in rollout length, problem difficulty and sample efficiency. Further, models with long chain-of-thought traces increase inference cost and latency, making efficiency critical for practical deployment. We present Apriel-Reasoner, trained with a fully reproducible multi-domain RL post-training recipe on Apriel-Base, a 15B-parameter open-weight LLM, across five domains using public datasets: mathematics, code generation, instruction following, logical puzzles and function calling. We introduce an adaptive domain sampling mechanism that preserves target domain ratios despite heterogeneous rollout dynamics, and a difficulty-aware extension of the standard length penalty that, with no additional training overhead, encourages longer reasoning for difficult problems and shorter traces for easy ones. Trained with a strict 16K-token output budget, Apriel-Reasoner generalizes to 32K tokens at inference and improves over Apriel-Base on AIME 2025, GPQA, MMLU-Pro, and LiveCodeBench while producing 30-50% shorter reasoning traces. It matches strong open-weight models of similar size at lower token cost, thereby pushing the Pareto frontier of accuracy versus token budget.

Oscar Clivio, Divyat Mahajan, Perouz Taslakian et al.

Integrating expert knowledge, e.g. from large language models, into causal discovery algorithms can be challenging when the knowledge is not guaranteed to be correct. Expert recommendations may contradict data-driven results, and their reliability can vary significantly depending on the domain or specific query. Existing methods based on soft constraints or inconsistencies in predicted causal relationships fail to account for these variations in expertise. To remedy this, we propose L2D-CD, a method for gauging the correctness of expert recommendations and optimally combining them with data-driven causal discovery results. By adapting learning-to-defer (L2D) algorithms for pairwise causal discovery (CD), we learn a deferral function that selects whether to rely on classical causal discovery methods using numerical data or expert recommendations based on textual meta-data. We evaluate L2D-CD on the canonical Tübingen pairs dataset and demonstrate its superior performance compared to both the causal discovery method and the expert used in isolation. Moreover, our approach identifies domains where the expert's performance is strong or weak. Finally, we outline a strategy for generalizing this approach to causal discovery on graphs with more than two variables, paving the way for further research in this area.

Benjamin Leblanc, Mathieu Bazinet, Nathaniel D'Amours et al.

Both PAC-Bayesian and Sample Compress learning frameworks are instrumental for deriving tight (non-vacuous) generalization bounds for neural networks. We leverage these results in a meta-learning scheme, relying on a hypernetwork that outputs the parameters of a downstream predictor from a dataset input. The originality of our approach lies in the investigated hypernetwork architectures that encode the dataset before decoding the parameters: (1) a PAC-Bayesian encoder that expresses a posterior distribution over a latent space, (2) a Sample Compress encoder that selects a small sample of the dataset input along with a message from a discrete set, and (3) a hybrid between both approaches motivated by a new Sample Compress theorem handling continuous messages. The latter theorem exploits the pivotal information transiting at the encoder-decoder junction in order to compute generalization guarantees for each downstream predictor obtained by our meta-learning scheme.

Vincent Zhihao Zheng, Étienne Marcotte, Arjun Ashok et al.

Context-aided forecasting (CAF) holds promise for integrating domain knowledge and forward-looking information, enabling AI systems to surpass traditional statistical methods. However, recent empirical studies reveal a puzzling gap: multimodal models often fail to outperform their unimodal counterparts. We hypothesize that this underperformance stems from poor context quality in existing datasets, as verification is challenging. To address these limitations, we introduce a semi-synthetic data augmentation method that generates contexts both descriptive of temporal dynamics and verifiably complementary to numerical histories. This approach enables massive-scale dataset creation, resulting in CAF-7M, a corpus of 7 million context-augmented time series windows, including a rigorously verified test set. We demonstrate that semi-synthetic pre-training transfers effectively to real-world evaluation, and show clear evidence of context utilization. Our results suggest that dataset quality, rather than architectural limitations, has been the primary bottleneck in context-aided forecasting.

Léo Boisvert, Abhay Puri, Chandra Kiran Reddy Evuru et al.

The practice of fine-tuning AI agents on data from their own interactions--such as web browsing or tool use--, while being a strong general recipe for improving agentic capabilities, also introduces a critical security vulnerability within the AI supply chain. In this work, we show that adversaries can easily poison the data collection pipeline to embed hard-to-detect backdoors that are triggerred by specific target phrases, such that when the agent encounters these triggers, it performs an unsafe or malicious action. We formalize and validate three realistic threat models targeting different layers of the supply chain: 1) direct poisoning of fine-tuning data, where an attacker controls a fraction of the training traces; 2) environmental poisoning, where malicious instructions are injected into webpages scraped or tools called while creating training data; and 3) supply chain poisoning, where a pre-backdoored base model is fine-tuned on clean data to improve its agentic capabilities. Our results are stark: by poisoning as few as 2% of the collected traces, an attacker can embed a backdoor causing an agent to leak confidential user information with over 80% success when a specific trigger is present. This vulnerability holds across all three threat models. Furthermore, we demonstrate that prominent safeguards, including two guardrail models and one weight-based defense, fail to detect or prevent the malicious behavior. These findings highlight an urgent threat to agentic AI development and underscore the critical need for rigorous security vetting of data collection processes and end-to-end model supply chains.

Arjun Ashok, Andrew Robert Williams, Vincent Zhihao Zheng et al.

Forecasting in real-world settings requires models to integrate not only historical data but also relevant contextual information, often available in textual form. While recent work has shown that large language models (LLMs) can be effective context-aided forecasters via naïve direct prompting, their full potential remains underexplored. We address this gap with 4 strategies, providing new insights into the zero-shot capabilities of LLMs in this setting. ReDP improves interpretability by eliciting explicit reasoning traces, allowing us to assess the model's reasoning over the context independently from its forecast accuracy. CorDP leverages LLMs solely to refine existing forecasts with context, enhancing their applicability in real-world forecasting pipelines. IC-DP proposes embedding historical examples of context-aided forecasting tasks in the prompt, substantially improving accuracy even for the largest models. Finally, RouteDP optimizes resource efficiency by using LLMs to estimate task difficulty, and routing the most challenging tasks to larger models. Evaluated on different kinds of context-aided forecasting tasks from the CiK benchmark, our strategies demonstrate distinct benefits over naïve prompting across LLMs of different sizes and families. These results open the door to further simple yet effective improvements in LLM-based context-aided forecasting.

Alexandre Drouin, Étienne Marcotte, Nicolas Chapados

The estimation of time-varying quantities is a fundamental component of decision making in fields such as healthcare and finance. However, the practical utility of such estimates is limited by how accurately they quantify predictive uncertainty. In this work, we address the problem of estimating the joint predictive distribution of high-dimensional multivariate time series. We propose a versatile method, based on the transformer architecture, that estimates joint distributions using an attention-based decoder that provably learns to mimic the properties of non-parametric copulas. The resulting model has several desirable properties: it can scale to hundreds of time series, supports both forecasting and interpolation, can handle unaligned and non-uniformly sampled data, and can seamlessly adapt to missing data during training. We demonstrate these properties empirically and show that our model produces state-of-the-art predictions on multiple real-world datasets.

Alexandre Lacoste, Evan David Sherwin, Hannah Kerner et al.

Recent progress in self-supervision shows that pre-training large neural networks on vast amounts of unsupervised data can lead to impressive increases in generalisation for downstream tasks. Such models, recently coined as foundation models, have been transformational to the field of natural language processing. While similar models have also been trained on large corpuses of images, they are not well suited for remote sensing data. To stimulate the development of foundation models for Earth monitoring, we propose to develop a new benchmark comprised of a variety of downstream tasks related to climate change. We believe that this can lead to substantial improvements in many existing applications and facilitate the development of new applications. This proposal is also a call for collaboration with the aim of developing a better evaluation process to mitigate potential downsides of foundation models for Earth monitoring.

Philippe Brouillard, Perouz Taslakian, Alexandre Lacoste et al.

Causal discovery from observational data is a challenging task that can only be solved up to a set of equivalent solutions, called an equivalence class. Such classes, which are often large in size, encode uncertainties about the orientation of some edges in the causal graph. In this work, we propose a new set of assumptions that constrain possible causal relationships based on the nature of variables, thus circumscribing the equivalence class. Namely, we introduce typed directed acyclic graphs, in which variable types are used to determine the validity of causal relationships. We demonstrate, both theoretically and empirically, that the proposed assumptions can result in significant gains in the identification of the causal graph. We also propose causal discovery algorithms that make use of these assumptions and demonstrate their benefits on simulated and pseudo-real data.

Xiang Zhang, Alexandre Drouin, Raymond Li

This article introduces byteSteady -- a fast model for classification using byte-level n-gram embeddings. byteSteady assumes that each input comes as a sequence of bytes. A representation vector is produced using the averaged embedding vectors of byte-level n-grams, with a pre-defined set of n. The hashing trick is used to reduce the number of embedding vectors. This input representation vector is then fed into a linear classifier. A straightforward application of byteSteady is text classification. We also apply byteSteady to one type of non-language data -- DNA sequences for gene classification. For both problems we achieved competitive classification results against strong baselines, suggesting that byteSteady can be applied to both language and non-language data. Furthermore, we find that simple compression using Huffman coding does not significantly impact the results, which offers an accuracy-speed trade-off previously unexplored in machine learning.

Gintare Karolina Dziugaite, Alexandre Drouin, Brady Neal et al.

One of the principal scientific challenges in deep learning is explaining generalization, i.e., why the particular way the community now trains networks to achieve small training error also leads to small error on held-out data from the same population. It is widely appreciated that some worst-case theories -- such as those based on the VC dimension of the class of predictors induced by modern neural network architectures -- are unable to explain empirical performance. A large volume of work aims to close this gap, primarily by developing bounds on generalization error, optimization error, and excess risk. When evaluated empirically, however, most of these bounds are numerically vacuous. Focusing on generalization bounds, this work addresses the question of how to evaluate such bounds empirically. Jiang et al. (2020) recently described a large-scale empirical study aimed at uncovering potential causal relationships between bounds/measures and generalization. Building on their study, we highlight where their proposed methods can obscure failures and successes of generalization measures in explaining generalization. We argue that generalization measures should instead be evaluated within the framework of distributional robustness.

Alexandre Lacoste, Pau Rodríguez, Frédéric Branchaud-Charron et al.

Progress in the field of machine learning has been fueled by the introduction of benchmark datasets pushing the limits of existing algorithms. Enabling the design of datasets to test specific properties and failure modes of learning algorithms is thus a problem of high interest, as it has a direct impact on innovation in the field. In this sense, we introduce Synbols -- Synthetic Symbols -- a tool for rapidly generating new datasets with a rich composition of latent features rendered in low resolution images. Synbols leverages the large amount of symbols available in the Unicode standard and the wide range of artistic font provided by the open font community. Our tool's high-level interface provides a language for rapidly generating new distributions on the latent features, including various types of textures and occlusions. To showcase the versatility of Synbols, we use it to dissect the limitations and flaws in standard learning algorithms in various learning setups including supervised learning, active learning, out of distribution generalization, unsupervised representation learning, and object counting.

Philippe Brouillard, Sébastien Lachapelle, Alexandre Lacoste et al.

Learning a causal directed acyclic graph from data is a challenging task that involves solving a combinatorial problem for which the solution is not always identifiable. A new line of work reformulates this problem as a continuous constrained optimization one, which is solved via the augmented Lagrangian method. However, most methods based on this idea do not make use of interventional data, which can significantly alleviate identifiability issues. This work constitutes a new step in this direction by proposing a theoretically-grounded method based on neural networks that can leverage interventional data. We illustrate the flexibility of the continuous-constrained framework by taking advantage of expressive neural architectures such as normalizing flows. We show that our approach compares favorably to the state of the art in a variety of settings, including perfect and imperfect interventions for which the targeted nodes may even be unknown.

Ulysse Côté-Allard, Cheikh Latyr Fall, Alexandre Drouin et al.

In recent years, deep learning algorithms have become increasingly more prominent for their unparalleled ability to automatically learn discriminant features from large amounts of data. However, within the field of electromyography-based gesture recognition, deep learning algorithms are seldom employed as they require an unreasonable amount of effort from a single person, to generate tens of thousands of examples. This work's hypothesis is that general, informative features can be learned from the large amounts of data generated by aggregating the signals of multiple users, thus reducing the recording burden while enhancing gesture recognition. Consequently, this paper proposes applying transfer learning on aggregated data from multiple users, while leveraging the capacity of deep learning algorithms to learn discriminant features from large datasets. Two datasets comprised of 19 and 17 able-bodied participants respectively (the first one is employed for pre-training) were recorded for this work, using the Myo Armband. A third Myo Armband dataset was taken from the NinaPro database and is comprised of 10 able-bodied participants. Three different deep learning networks employing three different modalities as input (raw EMG, Spectrograms and Continuous Wavelet Transform (CWT)) are tested on the second and third dataset. The proposed transfer learning scheme is shown to systematically and significantly enhance the performance for all three networks on the two datasets, achieving an offline accuracy of 98.31% for 7 gestures over 17 participants for the CWT-based ConvNet and 68.98% for 18 gestures over 10 participants for the raw EMG-based ConvNet. Finally, a use-case study employing eight able-bodied participants suggests that real-time feedback allows users to adapt their muscle activation strategy which reduces the degradation in accuracy normally experienced over time.

Alexandre Drouin, Toby Dylan Hocking, François Laviolette

Learning a regression function using censored or interval-valued output data is an important problem in fields such as genomics and medicine. The goal is to learn a real-valued prediction function, and the training output labels indicate an interval of possible values. Whereas most existing algorithms for this task are linear models, in this paper we investigate learning nonlinear tree models. We propose to learn a tree by minimizing a margin-based discriminative objective function, and we provide a dynamic programming algorithm for computing the optimal solution in log-linear time. We show empirically that this algorithm achieves state-of-the-art speed and prediction accuracy in a benchmark of several data sets.

Alexandre Drouin, Frédéric Raymond, Gaël Letarte St-Pierre et al.

Antimicrobial resistance is an important public health concern that has implications in the practice of medicine worldwide. Accurately predicting resistance phenotypes from genome sequences shows great promise in promoting better use of antimicrobial agents, by determining which antibiotics are likely to be effective in specific clinical cases. In healthcare, this would allow for the design of treatment plans tailored for specific individuals, likely resulting in better clinical outcomes for patients with bacterial infections. In this work, we present the recent work of Drouin et al. (2016) on using Set Covering Machines to learn highly interpretable models of antibiotic resistance and complement it by providing a large scale application of their method to the entire PATRIC database. We report prediction results for 36 new datasets and present the Kover AMR platform, a new web-based tool allowing the visualization and interpretation of the generated models.

Alexandre Drouin, Sébastien Giguère, Maxime Déraspe et al.

The Set Covering Machine (SCM) is a greedy learning algorithm that produces sparse classifiers. We extend the SCM for datasets that contain a huge number of features. The whole genetic material of living organisms is an example of such a case, where the number of feature exceeds 10^7. Three human pathogens were used to evaluate the performance of the SCM at predicting antimicrobial resistance. Our results show that the SCM compares favorably in terms of sparsity and accuracy against L1 and L2 regularized Support Vector Machines and CART decision trees. Moreover, the SCM was the only algorithm that could consider the full feature space. For all other algorithms, the latter had to be filtered as a preprocessing step.

Alexandre Drouin, Sébastien Giguère, Vladana Sagatovich et al.

The increased affordability of whole genome sequencing has motivated its use for phenotypic studies. We address the problem of learning interpretable models for discrete phenotypes from whole genomes. We propose a general approach that relies on the Set Covering Machine and a k-mer representation of the genomes. We show results for the problem of predicting the resistance of Pseudomonas Aeruginosa, an important human pathogen, against 4 antibiotics. Our results demonstrate that extremely sparse models which are biologically relevant can be learnt using this approach.

Sébastien Giguère, Mario Marchand, François Laviolette et al.

We propose a specialized string kernel for small bio-molecules, peptides and pseudo-sequences of binding interfaces. The kernel incorporates physico-chemical properties of amino acids and elegantly generalize eight kernels, such as the Oligo, the Weighted Degree, the Blended Spectrum, and the Radial Basis Function. We provide a low complexity dynamic programming algorithm for the exact computation of the kernel and a linear time algorithm for it's approximation. Combined with kernel ridge regression and SupCK, a novel binding pocket kernel, the proposed kernel yields biologically relevant and good prediction accuracy on the PepX database. For the first time, a machine learning predictor is capable of accurately predicting the binding affinity of any peptide to any protein. The method was also applied to both single-target and pan-specific Major Histocompatibility Complex class II benchmark datasets and three Quantitative Structure Affinity Model benchmark datasets. On all benchmarks, our method significantly (p-value < 0.057) outperforms the current state-of-the-art methods at predicting peptide-protein binding affinities. The proposed approach is flexible and can be applied to predict any quantitative biological activity. The method should be of value to a large segment of the research community with the potential to accelerate peptide-based drug and vaccine development.