archimedes-Artificial Intelligence, Data Science, Algorithms-greece

 
Artificial Intelligence
 
Data Science
 
Algorithms

The Underlying Logic of Language Models - Ryan Cotterell (ETH Zurich, Switzerland)

3000_followers___linkedin
Dates
2025-07-16 10:00 - 11:30
Venue
101, AUEB Troias Building (2 Troias Str., Troias wing, 1st floor) and virtually via Microsoft Teams (meeting information shown below)

 

Title: The Underlying Logic of Language Models

Speaker:Prof. Ryan Cotterell (ETH Zurich, Switzerland)

Abstract: The formal basis of the theory of computation lies in the study of languages, subsets of Σ*, the set of all strings over an alphabet Σ. Models of computation can be taxonomized into the languages they can decide on, i.e., which languages a model can be used to determine membership of. For instance, finite-state automata can decide membership in the regular languages. Language models are probabilistic generalizations of language where the notion of a set is relaxed into one of a probability distribution over Σ*. Recently, language models parameterized using recurrent neural networks, transformers, and state-space models have achieved enormous success in natural language processing. Similarly to how theorists have taxonomized models of deterministic computation, researchers have been made to taxonomize the expressivity of language models based on various architectures in terms of the distributions over strings they can represent. This tutorial presents a self-contained overview of the formal methods used to taxonomize the expressivity of language models, which encompass formal language and automata theory, various forms of formal logic, circuit complexity, and programming languages such as RASP. For example, we illustrate how transformers, under varying assumptions, can be characterized by different fragments of formal logic.

ryan
Short Biography: Ryan has been an assistant professor of computer science at ETH Zürich since 2020. Previously, he was a lecturer at the University of Cambridge. His PhD is from Johns Hopkins University, where he was advised by Jason Eisner. His research interests include natural language processing, computational linguistics, and machine learning. He has publishes at natural language processing venues (ACL, NAACL, EMNLP) venues as well as machine learning venues (NeurIPS, ICML, ICLR). We has additionally won various paper awards, including the overall best paper at ACL 2017.

Microsoft Teams 
Join the meeting now
Meeting ID: 392 124 408 383 7
Passcode: dP6gt3Ho
 
 

Vision

To position Greece as a leading player in AI and Data Science

image
image

Mission

To build an AI Excellence Hub in Greece where the international research community can connect, groundbreaking ideas can thrive, and the next generation of scientists emerges, shaping a brighter future for Greece and the world

 

Welcome to ARCHIMEDES, a vibrant research hub connecting the global AI and Data Science research community fostering groundbreaking research in Greece and beyond. Its dedicated core team, comprising lead researchers, affiliated researchers, Post-Docs, PhDs and interns, is committed to advancing basic and applied research in Artificial Intelligence and its supporting disciplines, including Algorithms, Statistics, Learning Theory, and Game Theory organized around 8 core research areas. By collaborating with Greek and Foreign Universities and Research Institutes, ARCHIMEDES disseminates its research findings fostering knowledge exchange and providing enriching opportunities for students. Leveraging AI to address real-world challenges, ARCHIMEDES promotes innovation within the Greek ecosystem and extends its societal impact. Established in January 2022, as a research unit of the Athena Research Center with support from the Committee Greece 2021, ARCHIMEDES is funded for its first four years by the EU Recovery and Resilience Facility (RRF).

 
 

NEWS

 
Archimedes Workshop on Dialect NLP

Archimedes Workshop on Dialect NLP

Upcoming workshop on Dialect NLP on “Standardization and Variation for Dialect Varieties with Universal Dependencies as an Application Framework” coming up. We are excited to announce that the Dialect NLP team at Archimedes, Athena Research Center, Greece, is organizing a workshop in collaboration with the MaiNLP Research Lab at Ludwig Maximilian University (LMU) of Munich. Workshop details:Workshop info: UniDive WG1 Workshop on Dialect Varieties Dates: September 5–6, 2025 Location: LMU Munich Funded by: UniDive – Universality, Diversity and Idiosyncrasy in Language Technology Archimedes: Dialect NLP and Linguistic Diversity At Archimedes, our Dialect NLP team focuses on the intersection of AI and linguistic diversity, with an emphasis on under-resourced and endangered language varieties, both Greek and non-Greek. Our research includes:   • Dialectal speech-to-text and morphosyntactic modeling  • Dialect-to-standard normalization for Greek varieties   • Intra-dialectal variation analysis   • A critical perspective on annotation frameworks like Universal Dependencies (UD), especially when applied to non-standard or endangered varieties Our resources cover:   • Standard Modern Greek: GUD Treebank   • Greek dialects: Cretan, Messinian, Lesbian, Cypriot, Griko/Greko, Aperathitika   • Non-Greek varieties spoken in Greece: Pomak, Arvanitika All datasets, models, and tools we develop are freely available to the research community. About Our Collaborators at MaiNLP (LMU Munich) The MaiNLP Research Lab at Ludwig-Maximilians-Universität Munich conducts research in Natural Language Processing, combining computer science, linguistics, and cognitive science. Their focus is on human-facing NLP: developing models that are robust to variation, fair, and reflective of human annotation diversity. MaiNLP is also leading the ERC Consolidator Grant project DIALECT, which explores natural language understanding for non-standard languages and dialects—making them an ideal partner in our shared mission to design NLP systems that work for all language varieties. Workshop Goals This workshop will:   • Document current practices in dialectal UD treebanks (e.g., orthographic/phonological variation, lemmatization, interference)   • Identify gaps in annotation and processing across dialects   • Develop shared methodologies for handling variation, improving consistency, and enabling knowledge transfer   • Foster community engagement with researchers working on under-resourced dialects and less-studied language families   • Ground discussions in theoretical insights from Plank (2016): What to do about non-standard language in NLP It is an opportunity to reflect on how annotation frameworks and tools serve—or fall short of serving— dialectal and endangered varieties, and how AI can support documentation, analysis, and preservation of linguistic diversity. We look forward to sharing our findings and engaging with colleagues working on inclusive, culturally informed, and variation-aware NLP. Stay tuned for more updates!

Archimedes and the Biomedical Research Foundation Share Latest Findings in AI and Medicine

Archimedes and the Biomedical Research Foundation Share Latest Findings in AI and Medicine

Archimedes and the Biomedical Research Foundation of the Academy of Athens successfully hosted a special collaborative session at the Panhellenic Working Group Seminars of the Hellenic Society of Cardiology. This session focused on innovative applications of artificial intelligence in medicine, with a particular emphasis on advancements in cardiology. Held on Friday, February 7, 2025, in Room MC2 of the Megaron Athens International Conference Centre, the event brought together leading experts to explore how artificial intelligence is transforming cardiovascular medicine.

Archimedes Reaches Milestone of 200 Publications

Archimedes Reaches Milestone of 200 Publications

Archimedes is proud to announce that its researchers have published over 200 scientific publications in top-tier conferences (NeurIPS, ICLR, ICML) and journals.Archimedes maintains a vibrant scientific community of over 130 researchers, including more than 60 senior researchers (faculty members from Greece and abroad), 12 postdoctoral fellows, and 55 PhD students, along with over 20 undergraduate interns from various disciplines.

 
 

The project “ARCHIMEDES Unit: Research in Artificial Intelligence, Data Science and Algorithms” with code OPS 5154714 is implemented by the National Recovery and Resilience Plan “Greece 2.0” and is funded by the European Union – NextGenerationEU.

greece2.0 eu_arch_logo_en

 

Stay connected! Subscribe to our mailing list by emailing sympa@lists.athenarc.gr
with the subject "subscribe archimedes-news Firstname LastName"
(replace with your details)