Tootfinder

Towards Language-Augmented Multi-Agent Deep Reinforcement Learning
Maxime Toquebiau, Jae-Yun Jun, Fa\"iz Benamar, Nicolas Bredeche
https://arxiv.org/abs/2506.05236

Towards Language-Augmented Multi-Agent Deep Reinforcement Learning
Communication is a fundamental aspect of coordinated behavior in multi-agent reinforcement learning. Yet, most prior works in this field have focused on emergent communication protocols developed from scratch, often resulting in inefficient or non-interpretable systems. Inspired by the role of language in natural intelligence, we investigate how grounding agents in a human-defined language can improve learning and coordination of multiple embodied agents. We propose a framework in which agents …

Trigonal Canonically Fibered Surfaces
Houari Benammar Ammar, Xi Chen, Nathan Grieve
https://arxiv.org/abs/2506.01526 https://arxiv.or…

Trigonal Canonically Fibered Surfaces
We fix some gaps of a proof of Xiao's conjecture on canonically fibered surfaces of relative genus 5 by the second author. Our argument simplifies the original proof and gives a much better bound on the geometric genus of the surface. Also we apply the same argument to canonically fibered surfaces of relative genus 3 and 4 to obtain some Noether-type inequalities for these surfaces.

Image-based physical characterization of magnetotactic bacteria from an environmental sample
Mara Smite, Mihails Birjukovs, Mila Sirinelli-Kojadinovic, Sandrine Grosse, Agnese Kokina, Janis Liepins, Dita Gudra, Megija Lunge, Davids Fridmanis, Mihaly Posfai, Andrejs Cebers, Damien Faivre, Guntars Kitenbergs
https://arxiv.org/abs/…

Image-based physical characterization of magnetotactic bacteria from an environmental sample
Magnetotactic bacteria (MTB) are a diverse group of microorganisms that are able to biomineralize magnetic nanoparticles. Most MTB remain uncultured, making population-level characterization from natural environments difficult. We report the discovery of a new and diverse MTB-rich site in the Ogre River, Latvia, and present an integrated approach combining 16S rRNA sequencing, transmission electron microscopy, and novel open-source, automated image-based physical methods to characterize bacteri…