Quentin Cappart
YOU?
Author Swipe
View article: Malice in Agentland: Down the Rabbit Hole of Backdoors in the AI Supply Chain
Malice in Agentland: Down the Rabbit Hole of Backdoors in the AI Supply Chain Open
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…
View article: DoomArena: A framework for Testing AI Agents Against Evolving Security Threats
DoomArena: A framework for Testing AI Agents Against Evolving Security Threats Open
We present DoomArena, a security evaluation framework for AI agents. DoomArena is designed on three principles: 1) It is a plug-in framework and integrates easily into realistic agentic frameworks like BrowserGym (for web agents) and $τ$-b…
View article: Learning Valid Dual Bounds in Constraint Programming: Boosted Lagrangian Decomposition with Self-Supervised Learning
Learning Valid Dual Bounds in Constraint Programming: Boosted Lagrangian Decomposition with Self-Supervised Learning Open
Lagrangian decomposition (LD) is a relaxation method that provides a dual bound for constrained optimization problems by decomposing them into more manageable sub-problems. This bound can be used in branch-and-bound algorithms to prune the…
View article: Improving Column Complementarity in a Restricted Master Heuristic with a Grasp-Guided Completion Application to the Vehicle Routing Problem with Stochastic Demands
Improving Column Complementarity in a Restricted Master Heuristic with a Grasp-Guided Completion Application to the Vehicle Routing Problem with Stochastic Demands Open
View article: The BrowserGym Ecosystem for Web Agent Research
The BrowserGym Ecosystem for Web Agent Research Open
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 …
View article: Learning and fine-tuning a generic value-selection heuristic inside a constraint programming solver
Learning and fine-tuning a generic value-selection heuristic inside a constraint programming solver Open
Constraint programming is known for being an efficient approach to solving combinatorial problems. Important design choices in a solver are the branching heuristics , designed to lead the search to the best solutions in a minimum amount of…
View article: Winning the 2023 CityLearn Challenge: A Community-Based Hierarchical Energy Systems Coordination Algorithm
Winning the 2023 CityLearn Challenge: A Community-Based Hierarchical Energy Systems Coordination Algorithm Open
The effective management and control of building energy systems are crucial for reducing the energy consumption peak loads, CO2 emissions, and ensuring the stability of the power grid, while maintaining optimal comfort levels within buildi…
View article: Learning Valid Dual Bounds in Constraint Programming: Boosted Lagrangian Decomposition with Self-Supervised Learning
Learning Valid Dual Bounds in Constraint Programming: Boosted Lagrangian Decomposition with Self-Supervised Learning Open
Lagrangian decomposition (LD) is a relaxation method that provides a dual bound for constrained optimization problems by decomposing them into more manageable sub-problems. This bound can be used in branch-and-bound algorithms to prune the…
View article: Boost Embodied AI Models with Robust Compression Boundary
Boost Embodied AI Models with Robust Compression Boundary Open
The rapid improvement of deep learning models with the integration of the physical world has dramatically improved embodied AI capabilities. Meanwhile, the powerful embodied AI models and their scales place an increasing burden on deployme…
View article: WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks
WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks Open
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 apply…
View article: WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?
WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks? Open
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 utili…
View article: Towards a Generic Representation of Combinatorial Problems for Learning-Based Approaches
Towards a Generic Representation of Combinatorial Problems for Learning-Based Approaches Open
In recent years, there has been a growing interest in using learning-based approaches for solving combinatorial problems, either in an end-to-end manner or in conjunction with traditional optimization algorithms. In both scenarios, the cha…
View article: Deep Learning for Data-Driven Districting-and-Routing
Deep Learning for Data-Driven Districting-and-Routing Open
Districting-and-routing is a strategic problem aiming to aggregate basic geographical units (e.g., zip codes) into delivery districts. Its goal is to minimize the expected long-term routing cost of performing deliveries in each district se…
View article: Learning Lagrangian Multipliers for the Travelling Salesman Problem
Learning Lagrangian Multipliers for the Travelling Salesman Problem Open
Lagrangian relaxation is a versatile mathematical technique employed to relax constraints in an optimization problem, enabling the generation of dual bounds to prove the optimality of feasible solutions and the design of efficient propagat…
View article: Global Rewards in Multi-Agent Deep Reinforcement Learning for Autonomous Mobility on Demand Systems
Global Rewards in Multi-Agent Deep Reinforcement Learning for Autonomous Mobility on Demand Systems Open
We study vehicle dispatching in autonomous mobility on demand (AMoD) systems, where a central operator assigns vehicles to customer requests or rejects these with the aim of maximizing its total profit. Recent approaches use multi-agent de…
View article: An Exact Framework for Solving the Space-Time Dependent TSP
An Exact Framework for Solving the Space-Time Dependent TSP Open
Many real-world scenarios involve solving bi-level optimization problems in which there is an outer discrete optimization problem, and an inner problem involving expensive or black-box computation. This arises in space-time dependent varia…
View article: Improved Peel-and-Bound: Methods for Generating Dual Bounds with Multivalued Decision Diagrams
Improved Peel-and-Bound: Methods for Generating Dual Bounds with Multivalued Decision Diagrams Open
Decision diagrams are an increasingly important tool in cutting-edge solvers for discrete optimization. However, the field of decision diagrams is relatively new, and is still incorporating the library of techniques that conventional solve…
View article: Dynamic Routing and Wavelength Assignment with Reinforcement Learning
Dynamic Routing and Wavelength Assignment with Reinforcement Learning Open
With the rapid developments in communication systems, and considering their dynamic nature, all-optical networks are becoming increasingly complex. This study proposes a novel method based on deep reinforcement learning for the routing and…
View article: Improved Peel-and-Bound: Methods for Generating Dual Bounds with Multivalued Decision Diagrams
Improved Peel-and-Bound: Methods for Generating Dual Bounds with Multivalued Decision Diagrams Open
Decision diagrams are an increasingly important tool in cutting-edge solvers for discrete optimization. However, the field of decision diagrams is relatively new, and is still incorporating the library of techniques that conventional solve…
View article: Learning a Generic Value-Selection Heuristic Inside a Constraint Programming Solver
Learning a Generic Value-Selection Heuristic Inside a Constraint Programming Solver Open
Constraint programming is known for being an efficient approach for solving combinatorial problems. Important design choices in a solver are the branching heuristics, which are designed to lead the search to the best solutions in a minimum…
View article: Peel-and-Bound: Generating Stronger Relaxed Bounds with Multivalued Decision Diagrams
Peel-and-Bound: Generating Stronger Relaxed Bounds with Multivalued Decision Diagrams Open
Decision diagrams are an increasingly important tool in cutting-edge solvers for discrete optimization. However, the field of decision diagrams is relatively new, and is still incorporating the library of techniques that conventional solve…
View article: Learning the travelling salesperson problem requires rethinking generalization
Learning the travelling salesperson problem requires rethinking generalization Open
End-to-end training of neural network solvers for graph combinatorial optimization problems such as the Travelling Salesperson Problem (TSP) have seen a surge of interest recently, but remain intractable and inefficient beyond graphs with …
View article: The Machine Learning for Combinatorial Optimization Competition (ML4CO):\n Results and Insights
The Machine Learning for Combinatorial Optimization Competition (ML4CO):\n Results and Insights Open
Combinatorial optimization is a well-established area in operations research\nand computer science. Until recently, its methods have focused on solving\nproblem instances in isolation, ignoring that they often stem from related data\ndistr…
View article: The Machine Learning for Combinatorial Optimization Competition (ML4CO): Results and Insights
The Machine Learning for Combinatorial Optimization Competition (ML4CO): Results and Insights Open
Combinatorial optimization is a well-established area in operations research and computer science. Until recently, its methods have focused on solving problem instances in isolation, ignoring that they often stem from related data distribu…
View article: Peel-And-Bound: Generating Stronger Relaxed Bounds with Multivalued Decision Diagrams
Peel-And-Bound: Generating Stronger Relaxed Bounds with Multivalued Decision Diagrams Open
Decision diagrams are an increasingly important tool in cutting-edge solvers for discrete optimization. However, the field of decision diagrams is relatively new, and is still incorporating the library of techniques that conventional solve…
View article: Efficient Minimum Weight Vertex Cover Heuristics Using Graph Neural Networks
Efficient Minimum Weight Vertex Cover Heuristics Using Graph Neural Networks Open
Minimum weighted vertex cover is the NP-hard graph problem of choosing a subset of vertices incident to all edges such that the sum of the weights of the chosen vertices is minimum. Previous efforts for solving this in practice have typica…
View article: On Causal Inference for Data-free Structured Pruning
On Causal Inference for Data-free Structured Pruning Open
Neural networks (NNs) are making a large impact both on research and industry. Nevertheless, as NNs' accuracy increases, it is followed by an expansion in their size, required number of compute operations and energy consumption. Increase i…
View article: Combinatorial Optimization and Reasoning with Graph Neural Networks
Combinatorial Optimization and Reasoning with Graph Neural Networks Open
Combinatorial optimization is a well-established area in operations research and computer science. Until recently, its methods have mostly focused on solving problem instances in isolation, ignoring the fact that they often stem from relat…
View article: Combining Reinforcement Learning and Constraint Programming for Combinatorial Optimization
Combining Reinforcement Learning and Constraint Programming for Combinatorial Optimization Open
Combinatorial optimization has found applications in numerous fields, from aerospace to transportation planning and economics. The goal is to find an optimal solution among a finite set of possibilities. The well-known challenge one faces …
View article: SeaPearl: A Constraint Programming Solver guided by Reinforcement\n Learning
SeaPearl: A Constraint Programming Solver guided by Reinforcement\n Learning Open
The design of efficient and generic algorithms for solving combinatorial\noptimization problems has been an active field of research for many years.\nStandard exact solving approaches are based on a clever and complete\nenumeration of the …