Tootfinder

mmBERT: A Modern Multilingual Encoder with Annealed Language Learning
Marc Marone, Orion Weller, William Fleshman, Eugene Yang, Dawn Lawrie, Benjamin Van Durme
https://arxiv.org/abs/2509.06888

mmBERT: A Modern Multilingual Encoder with Annealed Language Learning
Encoder-only languages models are frequently used for a variety of standard machine learning tasks, including classification and retrieval. However, there has been a lack of recent research for encoder models, especially with respect to multilingual models. We introduce mmBERT, an encoder-only language model pretrained on 3T tokens of multilingual text in over 1800 languages. To build mmBERT we introduce several novel elements, including an inverse mask ratio schedule and an inverse temperature…

How Important Is Preserving Democracy to Ukrainians? A Machine Learning Analysis of Democratic Attitudes: https://benborges.xyz/2025/10/08/how-important-is-preserving-democracy.html

Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples
Alexandra Souly, Javier Rando, Ed Chapman, Xander Davies, Burak Hasircioglu, Ezzeldin Shereen, Carlos Mougan, Vasilios Mavroudis, Erik Jones, Chris Hicks, Nicholas Carlini, Yarin Gal, Robert Kirk
https://arxiv.org/abs/2510.07192…

Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples
Poisoning attacks can compromise the safety of large language models (LLMs) by injecting malicious documents into their training data. Existing work has studied pretraining poisoning assuming adversaries control a percentage of the training corpus. However, for large models, even small percentages translate to impractically large amounts of data. This work demonstrates for the first time that poisoning attacks instead require a near-constant number of documents regardless of dataset size. We co…

MobRT: A Digital Twin-Based Framework for Scalable Learning in Mobile Manipulation
Yilin Mei, Peng Qiu, Wei Zhang, WenChao Zhang, Wenjie Song
https://arxiv.org/abs/2510.04592 ht…

MobRT: A Digital Twin-Based Framework for Scalable Learning in Mobile Manipulation
Recent advances in robotics have been largely driven by imitation learning, which depends critically on large-scale, high-quality demonstration data. However, collecting such data remains a significant challenge-particularly for mobile manipulators, which must coordinate base locomotion and arm manipulation in high-dimensional, dynamic, and partially observable environments. Consequently, most existing research remains focused on simpler tabletop scenarios, leaving mobile manipulation relativel…

"Waarom is van der Burg woordvoerder buitenlandse zaken en niet woordvoerder van de binnenvaart? Want het maakt kennelijk toch allemaal niks uit".
Aldus Paternotte.
Van Der Burg is not amused en moppert iets over grappen die niet op z'n plek zijn.
Maar Paternotte heeft wel een punt: bij Oekraine neemt de VVD *wel* het voortouw, dan kunnen we heel veel, maar bij Gaza zijn we ineens een hulpeloos klein landje.

Communication-Efficient Distributed Computing Through Combinatorial Multi-Access Models
Shanuja Sasi, Onur G\"unl\"u
https://arxiv.org/abs/2508.05426 https://

Communication-Efficient Distributed Computing Through Combinatorial Multi-Access Models
This paper explores the multi-access distributed computing (MADC) model, a novel distributed computing framework where mapper and reducer nodes are distinct entities. Unlike traditional MapReduce frameworks, MADC leverages coding-theoretic techniques to minimize communication overhead without necessitating file replication across mapper nodes. We introduce a new approach utilizing combinatorial designs, specifically t-designs, to construct efficient coding schemes that achieve a computation loa…

Language Bias in Information Retrieval: The Nature of the Beast and Mitigation Methods
Jinrui Yang, Fan Jiang, Timothy Baldwin
https://arxiv.org/abs/2509.06195 https://

Language Bias in Information Retrieval: The Nature of the Beast and Mitigation Methods
Language fairness in multilingual information retrieval (MLIR) systems is crucial for ensuring equitable access to information across diverse languages. This paper sheds light on the issue, based on the assumption that queries in different languages, but with identical semantics, should yield equivalent ranking lists when retrieving on the same multilingual documents. We evaluate the degree of fairness using both traditional retrieval methods, and a DPR neural ranker based on mBERT and XLM-R. A…

Spirits in the Stacks: A Ghostly Tour at the Congregational Library & Archives
https://ift.tt/mb4RdeS
ANN: New Book on Matriculture, Shamanism, and the Authority of Women: The Powers That Be Linnéa…
via Input 4 RELCFP

Dynamics of Infection Spread and Hotspot Growth in Bi-Pathogen Networks
Alyssa Yu, Laura P. Schaposnik
https://arxiv.org/abs/2509.03374 https://arxiv.org/p…

Dynamics of Infection Spread and Hotspot Growth in Bi-Pathogen Networks
Understanding the spatio-temporal evolution of epidemics with multiple pathogens requires not only new theoretical models but also careful analysis of their practical consequences. Building on the Multiplex Bi-Virus Reaction-Diffusion framework (MBRD) introduced in our companion paper, we investigate how the super-infection model (MBRD-SI) and the co-infection model (MBRD-CI) behave under different epidemiological and network conditions. Through numerical experiments, we study the effects of pa…

MMBERT: Scaled Mixture-of-Experts Multimodal BERT for Robust Chinese Hate Speech Detection under Cloaking Perturbations
Qiyao Xue, Yuchen Dou, Ryan Shi, Xiang Lorraine Li, Wei Gao
https://arxiv.org/abs/2508.00760

MMBERT: Scaled Mixture-of-Experts Multimodal BERT for Robust Chinese Hate Speech Detection under Cloaking Perturbations
Hate speech detection on Chinese social networks presents distinct challenges, particularly due to the widespread use of cloaking techniques designed to evade conventional text-based detection systems. Although large language models (LLMs) have recently improved hate speech detection capabilities, the majority of existing work has concentrated on English datasets, with limited attention given to multimodal strategies in the Chinese context. In this study, we propose MMBERT, a novel BERT-based m…