Program_SLTC2024

November 27-29, 2024

Day 1, November, 27

Oral presentations are given in the lecture hall, Ada Lovelace. The frame is 10 minutes for presentation and 5 minutes for discussion.

Coffee is served in nearby Ljusgården where also posters will be on display

Time Content Chair
     
13.00 Welcome Jenny
13.05 Invited speaker, Tiago Pimentel  
Duplicating Vocabularies to Analyse Generalisation in Language Models
14.00 Speech and dialogue Johan Boye
Staffan Larsson: Pre-Generative Conversational AI, or How I Learned to Stop Worrying and Use LLMs
Christina Tånnander, Jim O'Regan, Shivam Mehta, David House, Jonas Beskow and Jens Edlund: Continuous features in neural TTS
Nikolai Ilinykh and Sharid Loáiciga: How do images help coreference? A case study on the multimodal Tell-me-more dataset
15.00 Coffee
15.00 Poster session
Birger Moell and Fabian Farestam: Swedish Medical Benchmark, an evaluation framework for LLMs in the Swedish medical domain
Somayeh Jafaritazehjani and Johanna Björklund: An Extensible Framework for Real-Time Conversational Avatars
Dimitrios Kokkinakis, Daniel Wojahn and Johan Järlehed: Analyzing Segregation Discourse in Sweden: Technological Methods and Empirical Data
Daniel Tufvesson and Lars Ahrenberg: Speech Act Classification of Swedish Sentences: Bootstrapping an embedding-based neural classifier from a rule-based classifier
Julian Schlenker: On the Efficacy of Language Adapters for Cross-lingual Transfer in English-centric LLMs
Niklas Zechner: Identifying book reviews in Swedish newspapers
Nina Khairova, Andrea Galassi, Fabrizio Lo Scudo, Bogdan Ivasiuk and Ivan Redozub: Unsupervised Misinformation Detection in News Articles Using Semantic Similarity: A Case Study on the Russia-Ukraine War Dataset
16.00 Classification Thomas Vakili
Sara Stymne: Direct Speech Identification in Swedish Literature without Graphical Speech Marking
Maria Irena Szawerna, Simon Dobnik, Ricardo Muñoz Sánchez, and Elena Volodina: Swedish Learner Essays Revisited: Further Insights into Detecting Personal Information
Igor Ryazanov: Predicting Perceived Agency in News Headlines
17.30 Departure for Dinner at Flygvapenmuseet
Frimurarhotellet 17.30, Campus 17.40

Day 2, November, 28

Oral presentations are given in the lecture hall, Ada Lovelace. The frame is 10 minutes for presentation and 5 minutes for discussion.

Coffee is served in Ljusgården where also posters are shown.

Time Content Chair
     
09.00 Invited speaker: David Samuel Marco
Findings of the BabyLM Challenge.
10.00 Coffee
10.25 Text profiling Sara Stymne
Giorgia Albertin and Dimitrios Kokkinakis: Defining Cohesion Features in the Study of Discourse Properties in Cognitive Impairment
Magnus Ahltorp and Maria Skeppstedt: 1 1/2 years of developing Word Rain< /td>
11.00 Readability, text complexity Maria Skeppstedt
Wilgot Brissman and Arne Jönsson: Exploring and Analyzing Differences Across Levels of Readability in Easy-to-Read Text
Birger Moell, Fredrik Sand Aronsson, and Johan Boye: Can LLMs analyze language complexity?
Daniel Holmer and Arne Jönsson: Auxiliary Techniques to Help Readers Understand Texts
12.00 Lunch break
Lunch at personal expense
13.00 LLMs and Evaluation Kätriin Kukk
Richard Johansson: How Well Do Large Language Models Disambiguate Swedish Words?
Fredrik Malmberg, Anna Klezovich, and Jonas Beskow: Evaluating Sign Language Representation Learning with Large Language Models
Birger Moell, Gustav Eje Henter, and Jonas Beskow: Evaluating Large Language Models with Human Feedback: Establishing a Swedish Benchmark
14.00 Panel: Prospects for Research in Language Technology
Panelists: Alexandra Kafka Larsson (to be confirmed), Beata Megyesi, David Samuel, Hannah Devinney, Tiago Pimentel
15.00 Coffee
15.00 Poster session
Tosin Adewumi, Nudrat Habib, Lama Alkhaled, Elisa H. and Barney Smith: On the Limitations of Large Language Models (LLMs): False Attribution
Niklas Wretblad, Fredrik Gordh Riseby, Rahul Biswas, Amin Ahmadi and Oskar Holmström: Understanding the Effects of Noise in Text-to-SQL: An Examination of the BIRD-Bench Benchmark
Mattias Appelgren and Simon Dobnik: To Your Left: A Dataset and a Task of Spatial Perspective Coordination
Innokentii Smirnov: Morphological analysis and lemmatisation for Hittite
Niklas Deworetzki, Peter Ljunglöf, and Nicholas Smallbone: Towards an Algebraic Approach for Corpus Queries
Elias Salvador Smidt Torjani, Airidas Brikas, and Daniel Hardt: LLMs as Proxy Survey Participants with RAG
William Schill and György Kovács: Sentiment Analysis in Swedish text, using decoder-only generative large language models for Nordic languages
16.00 Formal languages and cryptography Jens Edlund
Beáta Megyesi: The DECRYPT Project: Cross-Disciplinary Research for Historical Cryptology
Aarne Ranta: Towards Multilingual Autoformalization and Informalization of Mathematics
16.30 Closing session TBD

Day 3, November, 29

The workshop schedules can be found at the workshop's web pages

Time Workshop Room
     
9.00 - 17 Computational Social Science and Language Technology Ada Lovelace
9.00 - 17 Swedish Workshop on Conversational Systems von Neumann
9.00 - 12 Applications of Universal Dependencies Alan Turing