ESSLLI 2026

All evening lectures take place in the room D-1122 (first floor, across the bridge).

Raffaella Bernardi (Free University of Bozen-Bolzano)

Short bio: Raffaella Bernardi is Associate Professor at the Faculty of Engineering, Free University of Bozen-Bolzano, Italy. She works on Computational Linguistics focusing on the interplay between Language and Reasoning as displayed during conversations and/or in multimodal contexts. She is part of the Association of Logic, Language and Information (FoLLI)  for which she has served in multiple roles (as Secretary, as  Logic and Language Area Chair (AC) of the ESSLLI Student Session 2001, local Program Chair (PC) of ESSLLI 2016, PC chair of ESSLLI 2021, etc.). She is an active member of the international Association of Computational Linguistics (ACL), serving as a Senior Chair, AC of *ACL conferences; she is also an active member of the Italian Association of Computational Linguistics (AILC) serving as PC chair, AC, member of the scientific committee of AILC Lectures. She has recently been the EU representative within the ACL Sponsorship Board. She has also been quite active in disseminating Computational Linguistics by means of teaching activities. She has long been the local coordinator, at unibz and then at unitn, of the Erasmus Mundus European Masters Programme in Language and Communication Technologies. She is an ELLIS (European Laboratory for Learning and Intelligent Systems) elected Fellow. Currently, she is serving as Director of the MSc in Computing for Data Science (Faculty of Engineering), member at-large of the ACL Management Board, and member of AILC Management Board.

DATE: August 4th, 2026

TOPIC: Large Language Models through the lens of Logic, Language and Information: An ESSLLI Journey

Slides: https://drive.google.com/file/d/1OTbzD-_c7Oo-z4SM-bGsu8gZYue7suRj/view?usp=drive_link 

I grew up within the Logic & Language tradition and later joined a Distributional Semantics research group, where I witnessed the rise of Large Language Models.
The goal of the Logic & Language tradition has long been to uncover the formal relationship between syntax and semantics in natural language. Central to this perspective are the notions of denotation—entities, events, and truth values—and the compositional construction of meaning representations, enabling formal reasoning.
Working on Distributional Semantics allowed me to approach language from a different perspective and to explore new dimensions of meaning. In particular, my research on visually grounded conversational systems highlighted the importance of grounding language in perception and using it to accomplish tasks through interaction. Rather than moving away from the questions raised by formal semantics, this experience brought me back to them from a different angle, especially to the role of reasoning.
In this talk, I will revisit these ideas through the lens of my own ESSLLI journey. Looking back at the evolution of both the field and my own perspective—from Logic and Language to Distributional Semantics, and from Distributional Semantics to Large Language Models—I will discuss where we stand today, what Large Language Models have taught us, and, perhaps more importantly, what they are still missing. I will conclude by sharing my perspective on the challenges that lie ahead and that I hope to see tackled.

 

Philippe Schlenker (Institut Jean-Nicod, CNRS)

Short bio: Philippe Schlenker is a senior researcher at CNRS (Institut Jean-Nicod, Paris) and, for 6 months in 2026-2027, a visitor at ZAS Berlin with the support of the Humboldt Foundation (2025 Research Prize). He was educated at École Normale Supérieure (Paris), and obtained a Ph.D. in linguistics from MIT, and a Ph.D. in philosophy from EHESS (Paris).  His early interests included semantics, pragmatics, the philosophy of language and philosophical logic. He has conducted research on indexicals and indirect discourse, intensional semantics, anaphora, presuppositions, as well as semantic paradoxes.

In more recent work, he has advocated a program of ’Super Semantics’ that seeks to expand the traditional frontiers of the field. He has investigated the semantics of sign languages, with special attention both to their logical structure and to the rich iconic means that interact with it. In order to have a point of comparison for these iconic phenomena, Philippe Schlenker has also investigated the logic and typology of gestures in spoken language. In collaborative work with ethologists and psycholinguists, he has laid the groundwork for an "animal linguistics" that seeks to apply the general methods of formal linguistics to diverse animal signals, from primates to birds and bees. He has also advocated the development of a detailed semantics for music, albeit one that is very different from linguistic semantics.

DATE: August 6th, 2026

TOPIC: Meaning after the Multimodal Revolution: Seven Things Signs and Gestures Teach Us about Meaning

Materials: https://lingbuzz.net/lingbuzz/010015

Slides: https://drive.google.com/file/d/1_QKLFEymd9gXSX-4CWegzQPi1ujuydm_/view?usp=drive_link

Following a "multimodal revolution" in the language sciences, meaning is now studied not just in speech but also in sign languages ('sign') and in gestures. Seven lessons can be drawn from recent research. (1) Sometimes sign makes visible key logical elements that are covert in speech. (2) Sign combines the same meaning operations as speech with a richer depictive component. (3) Sign combines the same kind of syntax as speech with a dedicated syntax for its depictive elements. (4) Even the purely depictive component of sign is linguistically structured. (5) Gestures that replace words share some non-trivial properties with signs—but unlike signs, they are not words. (6) Gestures that accompany words make backgrounded (non-assertive) contributions. (7) Several of these properties apply beyond signs and gestures, suggesting that they have a deeper cognitive source, not a narrowly linguistic one. These lessons show that human meaning results from both discrete, logical-like and gradient, iconic components, integrated through powerful cognitive mechanisms across sign, speech and gestures.

 

Larry Moss (Mathematics Department, Indiana University) - Dick Oehrle Memorial Lecture

Short bio: Larry Moss is Professor of Mathematics at Indiana University in the USA, where he also is Adjunct Professor of Computer Science, Linguistics, and Philosophy, and affiliated with Cognitive Science and with Computational Linguistics.   His main research interests are in logic, especially with applied areas close to computer science and linguistics.  He is a past president of FoLLI and a past Editor-in-Chief of JoLLi.  ESSLLI and similar schools are his favorite academic event.

DATE: August 10th, 2026

TOPIC: How to not give up on logic and semantics.

Since the late 2010's, machines can do logical inference in natural language at human level, if not better in some ways.   Thus, we have a watershed moment in the history of logic.  One might even think that we should give up on understanding human inference in language, either on idealized inference as we usually do in logic, or on actual human-level inference as we do in semantics.  This talk is for people who don't want to give up on logic and semantics as tools for understanding inference in language.

 

Jan Hajič (Charles University)

Short bio: Jan Hajič is a Czech computational linguist renowned for his foundational contributions to natural language processing, particularly in the development of linguistically rich treebanks and statistical machine translation. He is a professor at Charles University in Prague and a figure whose work bridges rigorous formal linguistics with practical, data-driven computational methods, embodying a persistent drive to give the Czech language and other morphologically complex tongues a robust presence in the digital age.

DATE: August 13th, 2026

TOPIC: TBD

Abstract: TBD