1 citations · 1 across the 8 of their papers we have counts for
14 papers
PatchBench: Measuring Collateral Damage in Activation Patching
Alexi Canesse, Mathis Le Bail, Maël Jenny +3
An LLM safety patch can pass a benchmark while still being a poor repair. This risk is especially acute for jailbreak repairs, where the goal is to correct a specific unsafe behavi…
RAISED: Self-Distillation for Robustness to Prompt Injection in LLM Agents
Mohamed Dhouib, Clement Elliker, Alexi Canesse +6
Tool-using language-model agents are vulnerable to indirect prompt injection because they must act on untrusted external content. Existing training-time defenses can reduce attack…
Reformulation-Contrastive Learning for Mixed Integer Programs
Ousema Bouaneni, Mathis Le Bail, Clément Elliker +2
Mixed-integer linear programs (MILP) model many real-world decision problems, motivating machine-learning methods that exploit recurring structure to accelerate MILP solving. MILPs…
Kernel Token Contradiction: a Fast and Principled Approach for LLM Claim Uncertainty Quantification
Jérémie Dentan, Alexi Canesse, Mahammed El Sharkawy +1
Claim-level Uncertainty Quantification (UQ) aims to mitigate the lack of reliability of Large Language Models (LLMs) by evaluating the factuality of each claim in their outputs. We…
HypNO: A Graph-Based Neural Operator with Physics-Informed Message Passing for Hyperbolic Conservation Laws
Dimitrije Ždrale, Cassie An Jeng, Katie Wang +3
We introduce HypNO, a graph-based neural operator for scalar hyperbolic conservation laws. HypNO operates directly on a space-time graph of finite-volume cells and uses adjacency-f…
Efficient Document Tampering Localization with Multi-Level Discrepancy Features and Unified DCT-Quantization Embedding
Mohamed Dhouib, Ye Zhu, Sonia Vanier +1
Localizing document tampering is extremely challenging, as manipulations are crafted to appear visually consistent and often leave only subtle traces that are nearly invisible to t…
RIDE: An Open Dataset and Benchmark for Train Delay Prediction
Clément Elliker, Mathis Le Bail, Clément Mantoux +2
Train delay prediction is an important problem for both passengers and railway operators, yet progress in the field remains difficult to assess due to the lack of standardized data…
Decoupling Communication from Policy: Robust MARL under Bandwidth Constraints
Alexi Canesse, Benoît Goupil, Jesse Read +1
Communication enables coordination in multi-agent reinforcement learning (MARL), but many real-world applications, e.g., search-and-rescue with drone swarms, operate under severe b…
Leveraging Contrastive Learning for a Similarity-Guided Tampered Document Data Generation Pipeline
Mohamed Dhouib, Davide Buscaldi, Sonia Vanier +1
Detecting tampered text in document images is a challenging task due to data scarcity. To address this, previous work has attempted to generate tampered documents using rule-based…
MUCH: A Multilingual Claim Hallucination Benchmark
Jérémie Dentan, Alexi Canesse, Davide Buscaldi +2
Claim-level Uncertainty Quantification (UQ) is a promising approach to mitigate the lack of reliability in Large Language Models (LLMs). We introduce MUCH, the first claim-level UQ…
Guess or Recall? Training CNNs to Classify and Localize Memorization in LLMs
Jérémie Dentan, Davide Buscaldi, Sonia Vanier
Verbatim memorization in Large Language Models (LLMs) is a multifaceted phenomenon involving distinct underlying mechanisms. We introduce a novel method to analyze the different fo…
Unveiling Decision-Making in LLMs for Text Classification : Extraction of influential and interpretable concepts with Sparse Autoencoders
Mathis Le Bail, Jérémie Dentan, Davide Buscaldi +1
Sparse Autoencoders (SAEs) have been successfully used to probe Large Language Models (LLMs) and extract interpretable concepts from their internal representations. These concepts…
PACT: Pruning and Clustering-Based Token Reduction for Faster Visual Language Models
Mohamed Dhouib, Davide Buscaldi, Sonia Vanier +1
Visual Language Models require substantial computational resources for inference due to the additional input tokens needed to represent visual information. However, these visual to…
Predicting memorization within Large Language Models fine-tuned for classification
Jérémie Dentan, Davide Buscaldi, Aymen Shabou +1
Large Language Models have received significant attention due to their abilities to solve a wide range of complex tasks. However these models memorize a significant proportion of t…