Language model services count, limit and bill text in tokens, but the number of tokens needed for the same content depends on the language and on the tokenizer. We measure this token ratio relative to English for the 23 non-English official languages of the European Union on nine tokenizers of 2026: six company tokeniz...
Thomas Serval· Zenodo (CERN European Organi...· 0 citations
Network Intrusion Detection Systems (NIDSs) based on machine learning have demonstrated strong performance against known attack patterns. However, their robustness against adaptive evasion strategies orchestrated by modern Large Language Models (LLMs) remains largely unexplored. This study presents a reproducible frame...
Locally run small language models enable generating SQL from natural language without sending the database schema to third-party servers. However, a model fine-tuned only to "translate the question into SQL" can also translate destructive or unauthorized commands embedded in the prompt. In this report we measure, under...
Oğuz Kağan Kahraman, Galip Savas Ilgi· Zenodo (CERN European Organi...· 0 citations
The findings support use of LLMs in endodontic diagnosis and treatment planning as adjunctive tools, but not without expert supervision, and performance was endpoint-specific and substantially reduced on integrated assessment and low-prevalence categories.
Mahmoud Torabinejad, P. Tordik, Mona Alenezi et al.· Journal of Endodontics· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
A ReAct-style agent accumulates observations, intermediate results, and failed attempts in a single context. As this context grows, stale and off-path information consumes its limited capacity and obscures relevant evidence, making focused reasoning increasingly difficult. We introduce TreeFlex, a framework in which th...
Dong-Zhuo Li, Fang-Zhou Cheng, Wei-Wei Li et al.· 0 citations
Linters are widely used in code maintenance and real-time IDE editing to surface lint alerts related to performance, style, error-prone, and other patterns in source files. However, existing code benchmarks do not explicitly evaluate linter-guided file refactoring, where models must reduce lint alerts in source files w...
Tong-Wei Deng, Jin-Yu Zhu, Yue Cao et al.· 0 citations
Large language models often produce fluent but factually incorrect output. Gray-box detectors identify such hallucinations from the token-probability statistics emitted during generation, avoiding the need for white-box access to hidden states that frontier models do not provide. These detectors, however, typically rel...
S. Goldshtein, Yitzchak Grunbaum, Fabrizio Frasca et al.· 0 citations
Reasoning over infinitely long documents is becoming a reality with recurrent memory agents. However, increasing context length exacerbates the trade-off between reasoning performance and inference efficiency. Inspired by Kahneman's account of fast and slow thinking, we propose MemDuet, a dual-process memory framework...
Jiang-Cheng Song, Hao-Dong Jing, Yu-Ling Mu et al.· 0 citations
Interpreting an ambiguous utterance in real-world interactions often requires context beyond a single utterance. When the utterance refers to objects in a visual scene, it is necessary to maintain and update the set of plausible referents as context accumulates, without assuming that the available information identifie...
Shun Inadumi, Lee Sangmyeong, Koichiro Yoshino· 0 citations
Block-diffusion language models denoise a whole block of token positions in parallel so each forward pass can emit many tokens. However, blocks are still denoised in strict sequence, a cost that within-block acceleration leaves untouched. Here, we investigate decoding blocks concurrently and propose One Block Ahead (OB...
Yang-Yi Shen, Pin-Jui Ku, Nicolai Oswald et al.· 0 citations
The scaling relationship of compute and performance for language models (LMs) has been extensively studied; more recently, models are being designed to scale memory use, as well. In particular, adding embeddings for n-grams scales memory separately from compute use, because each pass only processes individual embedding...
Ian Magnusson, Noah A. Smith, P. W. Koh· 0 citations
Probes on activation vectors are widely used to detect misbehavior in language models, but their success relies on the assumption that models cannot control what their activations encode. Humans can learn, through neurofeedback, to control the firing of even single neurons, and recent studies investigated whether this...
Davide Baldelli, Robyn Kozierok, Mirko Bronzi et al.· 0 citations
A new computational approach reveals how cells communicate to assemble into tissues. This could accelerate stem cell engineering, improve disease models, and inform therapies.
Microsoft Research Blog· microsoft.comOct 15, 2026
Zero-knowledge proofs can verify computing results without rerunning workloads or exposing sensitive data. Magnetar helps bring them into the mainstream with support for standard programming languages, major speed gains, and quantum-resistant security. The post Proof-based general-purpose computing with Magnetar appeared first on Microsoft Research.
Learn how Google DeepMind and GovTech Singapore uses lightweight on-device Gemma models to intercept conversational scams in private chats with zero cloud data egress.
Microsoft Research Blog· microsoft.comOct 12, 2026
StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics. The post Simulating student learning for better AI tutoring appeared first on Microsoft Research.
