Portrait de Laurent Charlin

Laurent Charlin

Membre académique principal
Chaire en IA Canada-CIFAR
Professeur agrégé, HEC Montréal
Professeur associé, Université de Montréal, Département d'informatique et de recherche opérationnelle
Sujets de recherche
Apprentissage de représentations
Apprentissage par renforcement
Apprentissage profond
Exploration des données
IA pour la science
Modèles génératifs
Modèles probabilistes
Recherche d'information
Réseaux de neurones en graphes
Systèmes de recommandation
Traitement du langage naturel

Biographie

Laurent Charlin est titulaire d’une chaire en IA Canada-CIFAR et professeur agrégé à HEC Montréal. Il est également membre académique principal à Mila – Institut québécois d’intelligence artificielle.

Ses recherches portent sur le développement de nouveaux modèles d'apprentissage automatique pour aider à la prise de décision. Ses travaux récents concernent l'apprentissage à partir de données qui évoluent dans le temps. Il travaille également sur des applications dans des domaines tels que les systèmes de recommandation et l'optimisation.

Il est l'auteur de publications très citées sur les systèmes de dialogue (chatbots). Laurent Charlin a codéveloppé le Toronto Paper Matching System (TPMS), qui a été largement utilisé dans les conférences d'informatique pour faire correspondre les réviseur·euse·s aux articles. Il a également contribué à plusieurs MOOC récents, et a donné des conférences d'introduction et des interviews dans les médias pour contribuer au transfert de connaissances et améliorer la culture de l'IA.

Étudiants actuels

Publications

Discovering Data Structures: Nearest Neighbor Search and Beyond
Shivam Garg
Vatsal Sharan
Gregory Valiant
We propose a general framework for end-to-end learning of data structures. Our framework adapts to the underlying data distribution and prov… (voir plus)ides fine-grained control over query and space complexity. Crucially, the data structure is learned from scratch, and does not require careful initialization or seeding with candidate data structures/algorithms. We first apply this framework to the problem of nearest neighbor search. In several settings, we are able to reverse-engineer the learned data structures and query algorithms. For 1D nearest neighbor search, the model discovers optimal distribution (in)dependent algorithms such as binary search and variants of interpolation search. In higher dimensions, the model learns solutions that resemble k-d trees in some regimes, while in others, they have elements of locality-sensitive hashing. The model can also learn useful representations of high-dimensional data and exploit them to design effective data structures. We also adapt our framework to the problem of estimating frequencies over a data stream, and believe it could also be a powerful discovery tool for new problems.
How to Train Your LLM Web Agent: A Statistical Diagnosis
LLM-based web agents have recently made significant progress, but much of it has occurred in closed-source systems, widening the gap with op… (voir plus)en-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.
Addressing Concept Mislabeling in Concept Bottleneck Models Through Preference Optimization
Tianyue H. Zhang
Mateo Espinosa Zarlenga
Concept Bottleneck Models (CBMs) propose to enhance the trustworthiness of AI systems by constraining their decisions on a set of human-unde… (voir plus)rstandable concepts. However, CBMs typically assume that datasets contain accurate concept labels-an assumption often violated in practice, which we show can significantly degrade performance (by 25% in some cases). To address this, we introduce the Concept Preference Optimization (CPO) objective, a new loss function based on Direct Preference Optimization, which effectively mitigates the negative impact of concept mislabeling on CBM performance. We provide an analysis of key properties of the CPO objective, showing it directly optimizes for the concept's posterior distribution, and contrast it against Binary Cross Entropy (BCE), demonstrating that CPO is inherently less sensitive to concept noise. We empirically confirm our analysis by finding that CPO consistently outperforms BCE on three real-world datasets, both with and without added label noise. We make our code available on Github.
AInstein: Can AI Rediscover Scientific Concepts from First Principles?
Shambhavi Mishra
Jose Dolz
Christopher Pal
Large language models have demonstrated remarkable capabilities across diverse tasks, yet a fundamental question remains: can these models g… (voir plus)enuinely rediscover complex scientific insights, or do they merely recite memorized information? We present AInstein, a novel framework for evaluating whether language models can derive established scientific concepts from first principles when stripped of domain-specific terminology. Rather than testing the recall of scientific facts, we reformulate landmark discoveries as conceptual puzzles, challenging models to reconstruct the underlying technical solutions independently.
Evaluating and Improving LitLLMs with Deep Research
Issam Hadj Laradji
Krishnamurthy Dj Dvijotham
Jason Stanley
Christopher Pal
Literature reviews are an essential component of scientific research, but they remain time-intensive and challenging to write, especially du… (voir plus)e to the recent influx of research papers. This paper explores the zero-shot abilities of recent Large Language Models (LLMs) in assisting with the writing of literature reviews based on an abstract. We decompose the task into two components: (1) Retrieving related works given a query abstract and (2) Writing a literature review based on the retrieved results. We analyze how effective LLMs are for both components. For retrieval, we introduce a novel two-step search strategy that first uses an LLM to extract meaningful keywords from the abstract of a paper and then retrieves potentially relevant papers by querying an external knowledge base. Additionally, we study a prompting-based re-ranking mechanism with attribution and show that re-ranking doubles the normalized recall compared to naive search methods while providing insights into the LLM's decision-making process. In the generation phase, we propose a two-step approach that first outlines a plan for the review and then executes steps in the plan to generate the actual review. To evaluate different LLM-based literature review methods, we create test sets from arXiv papers using a protocol designed for rolling use with newly released LLMs to avoid test set contamination in zero-shot evaluations. We release this evaluation protocol to promote additional research and development in this regard. Our empirical results suggest that LLMs show promising potential for writing literature reviews when the task is decomposed into smaller components of retrieval and planning. Particularly, our ``Deep Research" retrieval variant improves coverage by over 5x compared to standard keyword search, addressing a key bottleneck in the pipeline. Further, we demonstrate that our planning-based approach achieves higher-quality reviews by minimizing hallucinated references in the generated review by 18-26\% compared to existing simpler LLM-based generation methods.
Towards sustainable energy use: Reinforcement learning for demand response in commercial buildings
Seyyedreza Madani
Pierre‐Olivier Pineau
Ysaël Desage
Demand response (DR) is a crucial strategy for balancing electricity demand and reducing environmental impact, especially as power consumpti… (voir plus)on continues to rise. Small and medium-sized commercial buildings hold significant potential for DR implementation due to their widespread presence and substantial contribution to overall energy use. This study proposes a novel framework to optimize energy management in these buildings by considering three types of loads: non-controllable, controllable with discrete action spaces (HVAC), and controllable with continuous action spaces (lighting). The objective is to minimize costs, reduce CO 2 emissions, improve occupant comfort, and shave peak loads as a unified goal. State-of-the-art reinforcement learning (RL) algorithms are employed and compared against traditional heuristic methods using real-world data. Results show that RL-based approaches can significantly lower energy costs and environmental impacts while maintaining occupant comfort, even under varying outdoor temperature conditions. By incorporating risk assessments through Value at Risk (VaR) and Conditional Value at Risk (CVaR) metrics, this study offers a robust solution for sustainable energy management, providing insights for policymakers and industry practitioners aiming for a more resilient energy future.
LLMs for Literature Review: Are we there yet?
Issam Hadj Laradji
Krishnamurthy Dj Dvijotham
Jason Stanley
Christopher Pal
Literature reviews are an essential component of scientific research, but they remain time-intensive and challenging to write, especially du… (voir plus)e to the recent influx of research papers. This paper explores the zero-shot abilities of recent Large Language Models (LLMs) in assisting with the writing of literature reviews based on an abstract. We decompose the task into two components: 1. Retrieving related works given a query abstract, and 2. Writing a literature review based on the retrieved results. We analyze how effective LLMs are for both components. For retrieval, we introduce a novel two-step search strategy that first uses an LLM to extract meaningful keywords from the abstract of a paper and then retrieves potentially relevant papers by querying an external knowledge base. Additionally, we study a prompting-based re-ranking mechanism with attribution and show that re-ranking doubles the normalized recall compared to naive search methods, while providing insights into the LLM's decision-making process. In the generation phase, we propose a two-step approach that first outlines a plan for the review and then executes steps in the plan to generate the actual review. To evaluate different LLM-based literature review methods, we create test sets from arXiv papers using a protocol designed for rolling use with newly released LLMs to avoid test set contamination in zero-shot evaluations. We release this evaluation protocol to promote additional research and development in this regard. Our empirical results suggest that LLMs show promising potential for writing literature reviews when the task is decomposed into smaller components of retrieval and planning. Further, we demonstrate that our planning-based approach achieves higher-quality reviews by minimizing hallucinated references in the generated review by 18-26% compared to existing simpler LLM-based generation methods.
PREFERENCE OPTIMIZATION FOR CONCEPT BOTTLENECK MODELS
Tianyue H. Zhang
Mateo Espinosa Zarlenga
Concept Bottleneck Models (CBMs) propose to enhance the trustworthiness of AI systems by constraining their decisions on a set of human-unde… (voir plus)rstandable concepts. However, CBMs typically assume that datasets contain accurate concept labels—an assumption often violated in practice, which we show can significantly degrade performance (by 25% in some cases). To address this, we introduce the Concept Preference Optimization (CPO) objective, a new loss function based on Direct Preference Optimization, which effectively mitigates the negative impact of concept mislabeling on CBM performance. We provide an analysis of some key properties of the CPO objective showing it directly optimizes for the concept’s posterior distribution, and contrast it against Binary Cross Entropy (BCE) where we show CPO is inherently less sensitive to concept noise. We empirically confirm our analysis finding that CPO consistently outperforms BCE in three real-world datasets with and without added label noise.
Integrating Present and Past in Unsupervised Continual Learning
Richard Zemel
Mengye Ren
We formulate a unifying framework for *unsupervised continual learning (UCL)*, which disentangles learning objectives that are specific to t… (voir plus)he present and the past data, encompassing *stability*, *plasticity*, and *cross-task consolidation*. The framework reveals that many existing UCL approaches overlook cross-task consolidation and try to balance plasticity and stability in a shared embedding space. This results in worse performance due to a lack of within-task data diversity and reduced effectiveness in learning the current task. Our method, *Osiris*, which explicitly optimizes all three objectives on separate embedding spaces, achieves state-of-the-art performance on all benchmarks, including two novel ones proposed in this paper featuring semantically structured task sequences. Finally, we show some preliminary evidence that continual models can benefit from such more realistic learning scenarios.
TEARS: Text Representations for Scrutable Recommendations.
Traditional recommender systems rely on high-dimensional (latent) embeddings for modeling user-item interactions, often resulting in opaque … (voir plus)representations that lack interpretability. Moreover, these systems offer limited control to users over their recommendations. Inspired by recent work, we introduce TExtuAl Representations for Scrutable recommendations (TEARS) to address these challenges. Instead of representing a user's interests through a latent embedding, TEARS encodes them in natural text, providing transparency and allowing users to edit them. To do so, TEARS uses a modern LLM to generate user summaries based on user preferences. Using these summaries, we take a hybrid approach where we use an optimal transport procedure to align the summaries' representation with the learned representation of a standard VAE for collaborative filtering. We find this approach can surpass the performance of popular VAE models while providing user-controllable recommendations. We also analyze the controllability of TEARS through three simulated user tasks to evaluate the effectiveness of a user editing its summary. A more detailed version of this manuscript with more experiments, baselines and detail is provided on arXiv.
Audio Prototypical Network For Controllable Music Recommendation
Traditional recommendation systems represent user preferences in dense representations obtained through black-box encoder models. While thes… (voir plus)e models often provide strong recommendation performance, they lack interpretability for users, leaving users unable to understand or control the system's modeling of their preferences. This limitation is especially challenging in music recommendation, where user preferences are highly personal and often evolve based on nuanced qualities like mood, genre, tempo, or instrumentation. In this paper, we propose an audio prototypical network for controllable music recommendation. This network expresses user preferences in terms of prototypes representative of semantically meaningful features pertaining to musical qualities. We show that the model obtains competitive recommendation performance compared to popular baseline models while also providing interpretable and controllable user profiles.
TEARS: Textual Representations for Scrutable Recommendations
Traditional recommender systems rely on high-dimensional (latent) embeddings for modeling user-item interactions, often resulting in opaque … (voir plus)representations that lack interpretability. Moreover, these systems offer limited control to users over their recommendations. Inspired by recent work, we introduce TExtuAl Representations for Scrutable recommendations (TEARS) to address these challenges. Instead of representing a user's interests through a latent embedding, TEARS encodes them in natural text, providing transparency and allowing users to edit them. To do so, TEARS uses a modern LLM to generate user summaries based on user preferences. We find the summaries capture user preferences uniquely. Using these summaries, we take a hybrid approach where we use an optimal transport procedure to align the summaries' representation with the learned representation of a standard VAE for collaborative filtering. We find this approach can surpass the performance of three popular VAE models while providing user-controllable recommendations. We also analyze the controllability of TEARS through three simulated user tasks to evaluate the effectiveness of a user editing its summary.