A search program saves work by shortcuts that stop a loop over candidates, or skip a candidate, where the choice cannot change. In the method described here the work of a search is spread over small sub-specifications of a logic with a decision procedure, and a shortcut acts only if a statement about its executed text,...
Taumorrow· Zenodo (CERN European Organi...· 1 citation
Root cause analysis (RCA) for incidents in large-scale cloud systems is a complex, knowledge-intensive task that often requires significant manual effort from on-call engineers (OCEs). Improving RCA is vital for accelerating the incident resolution process and reducing service downtime and manual efforts. Recent advanc...
Drishti Goel, Raghav Magazine, Supriyo Ghosh et al.· IEEE/ACM International Confe...· 1 citation
In domain-specific knowledge question answering (QA) for fields such as oceanography, systems face not only deployment constraints--offline operation, data sovereignty, and long-term cost control--but also a deeper, often overlooked problem: generative large language models (LLMs), as pure text generators, are architec...
Yiping Bai· Zenodo (CERN European Organi...· 0 citations
This report documents the ongoing development and empirical analysis of a 4.2-million-parameter Transformer-based small language model (SLM) trained from scratch by Novum Labs. The objective of this project is to evaluate the learning trajectory, convergence dynamics, semantic coherence, and dataset scaling characteris...
Khandakar Aaditta Arnab· Zenodo (CERN European Organi...· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Automatic disfluency correction is an important task in natural language processing, particularly for morphologically rich and low-resource languages such as Tamil. Disfluencies including repetitions, interjections, prolongations, and part-word repetitions can negatively affect the performance of speech transcription...
Open educational resources for two four-hour workshops in a master's programme in network forensics. In both workshops students build something small with the help of an AI assistant and then establish how far it can be trusted. The code is short; the work is in the checking. Workshop 1: AI-assisted network detector. S...
Olga Torstensson· Zenodo (CERN European Organi...· 0 citations
Foundations of Computational Linguistics: For EFL University Students is a free undergraduate textbook on precise representations and computational analysis of language. Part of the Theoretical Linguistics branch of English Language Studies for EFL University Students, it is designed for adult EFL students who have com...
Munassir Alhamami· Zenodo (CERN European Organi...· 0 citations
When a language model is allowed to revise a program repeatedly and a gate decides which revisions to keep, what the gate checks determines what the model does. We report a preregistered experiment in which an LLM optimiser rewrote a small posture-risk classifier for 20 revisions per trajectory under four acceptance ar...
Nicholas J. Lamb· Zenodo (CERN European Organi...· 0 citations
v4.5.1 is a correction release. The claims register's τ-phase matcher now recognises '46.0 µs' and no longer matches unrelated 46 µs delays, and its finding states the phase definition as the code computes it. In the console, a document pasted while OSIRIS was answering could be classed as typed and trained on; held me...
Devin Phillip Davis· Zenodo (CERN European Organi...· 0 citations
This preprint establishes an unconditional lower bound for complementing two-way deterministic finite automata. For every n ≥ 22, it constructs a language K_n over a fixed three-letter alphabet whose source state complexity is exactly n and whose complement state complexity is exactly 2n. These exact counts use accepta...
Sudais Arif· Zenodo (CERN European Organi...· 0 citations
Technical report on Linda-Pro 1.3.1: a local ensemble detector of AI-generated text in English, Polish and Russian (stylometry + two merged transformer models, one decision rule per language). Version 1.3.1 is a weights update retrained without text generated by Llama 3.x models, with additional permissively licensed h...
Vladyslav Manzyuk· Zenodo (CERN European Organi...· 0 citations
Verification code for the paper Self-audit horizons of finitely axiomatized theories: non-accumulation of finite reflection and the cost of external verification (H. Sendo). The code reproduces every numerical result of the paper: exact counts of formulas and the singularity analysis for first-order languages L(c,k,m)...
Hazime Sendo· Zenodo (CERN European Organi...· 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.
