Aalto computer scientists in EMNLP 2024
Empirical Methods in Natural Language Processing (EMNLP) is one of the leading conferences in the area of natural language processing and artificial intelligence. This year 1271 papers were accepted for the main conference and 1029 papers as Findings of EMNLP.
The conference is organised on 12-16 November 2024 in Miami, Florida.
Accepted papers
Click the title to see the authors and the abstract. Link to the paper open on different website.
Authors
Sam Spilsbury, Pekka Marttinen, Aleksander Ilin
Abstract
In-Context-learning and few-shot prompting are viable methods compositional output generation. However, these methods can be very sensitive to the choice of support examples used. Retrieving good supports from the training data for a given test query is already a difficult problem, but in some cases solving this may not even be enough. We consider the setting of grounded language learning problems where finding relevant supports in the same or similar states as the query may be difficult. We design an agent which instead generates possible supports inputs and targets current state of the world, then uses them in-context-learning to solve the test query. We show substantially improved performance on a previously unsolved compositional generalization test without a loss of performance in other areas. The approach is general and can even scale to instructions expressed in natural language.
Read more news
Samu Taulu develops next-generation brain imaging methods
Associate Professor of neurostimulation and neuroimaging Samu Taulu develops more precise methods for understanding the brain
EAA Best Paper and Presentation Award for Young Researchers for postdoctoral researcher Thomas Deppisch
The awarded work shows how speech can be made clearer in noisy environments while still preserving directional cues
Professor Hanna Maylett: 鈥淲ithout time spent together, a shared vision cannot emerge鈥
According to Hanna Maylett, associate professor of film directing, a director doesn鈥檛 need to know all the answers; rather, their role is to create a process through which those answers can be discovered together.