NLPitch

Every 2 weeks we hold a NLPitch session in which members of the ILLC can present their work to their colleagues. These sessions are usually held on Tuesday from 16:00 to 17:00, and organised by Sandro Pezzelle and Jaap Jumelet.

Previous speakers include:

• Mario Giulianelli - Analysing Human Strategies of Information Transmission as a Function of Discourse Context (slides)
• Rochelle Choenni - Stepmothers are mean and academics are pretentious: What do pretrained language models learn about you?
• Tomáš Musil (Charles University) - Embeddings, independent components, and a bit of philosophy of language
• Oskar van der Wal - Understanding the learning dynamics of undesirable bias in language models (slides)
• Aditya K Surikuchi - Visual Storytelling
• Jaap Jumelet - Feature interactions uncover the grammatical knowledge of neural models (slides)
• Mario Giulianelli and Jane Sinclair - GenBench, a benchmark for evaluating generalisation
• Bryan Eikema - Sampling from EBMs for Controlled Text Generation with Quality Estimates
• Ece Takmaz - The CMCL shared task on Multilingual and crosslingual prediction of human reading behavior
• Nicolò Brandizi - Emergent ImageNation
• Marek Holan (Palacký University Olomouc) - Extracting features from film-related data
• Alexandre Kabbach (University of Geneva) - Meaning in computational linguistics: from operationalization to objectivization
• Marianne de Heer Kloots - Relating deep neural networks and language models to human brain activations
• Joris Baan - Stop Measuring Calibration When Humans Disagree
• Alina Leidinger - Extracting stereotypes from implicitly hateful posts
• Esam Ghaleb - Multimodal Emotion Recognition through Deep Learning
• Michael Hanna - Subject-verb agreement in BERT: A case study in the functional relevance of probed information
• EMNLP 2022 edition: Jaap Jumelet, Mario Giulianelli, Joris Baan, Bryan Eikema