From the 1 of 3 linked papers with an AI index.
3 papers
Even More Deception: Objective Misalignment in Mixed-Motive LLM Multi-Agent Systems
Marylou Fauchard, Florian Carichon, Margarida Carvalho +1
The paper studies how misaligned objectives in large‑language‑model powered multi‑agent systems affect performance in the social deduction game Werewolf, showing that hidden object…
Revisiting the Neural Tangent Kernel: the role of large width and depth
William St-Arnaud, Margarida Carvalho, Golnoosh Farnadi
Overparameterized fully-connected neural networks have been shown to behave like kernel models when trained with gradient descent, assuming standard scaling conditions on the width…
Reasoning with Preference Constraints: A Benchmark for Language Models in Many-to-One Matching Markets
Marylou Fauchard, Florian Carichon, Margarida Carvalho +1
Recent advances in reasoning with large language models (LLMs) have demonstrated strong performance on complex mathematical tasks, including combinatorial optimization. Techniques…