https://vernadankers.com/blog/ Recent content in News on Verna Dankers Hugo -- gohugo.io en © Verna Dankers Thu, 07 May 2026 00:00:00 +0000 https://vernadankers.com/blog/news-2026-05/ Thu, 07 May 2026 00:00:00 +0000 https://vernadankers.com/blog/news-2026-05/ Presenting "Hidden in plain sight: injecting LLM memories for evaluation and safeguarding" at Utrecht University, U-NLP seminar. In this talk, I focus on a memory injection paradigm and present two use cases. Firstly, I introduce our new unlearning testbed, LACUNA, which injects synthetic personally identifiable information (PII) into known parameters to evaluate whether unlearning methods actually target the parameters that store the memories they aim to unlearn. We use LACUNA to demonstrate that state-of-the-art unlearning methods do not target the correct parameters, raising questions about whether they truly unlearn or merely obfuscate PII. https://vernadankers.com/blog/news-2026-03/ Sun, 01 Mar 2026 00:00:00 +0000 https://vernadankers.com/blog/news-2026-03/ Honored to receive the IVADO postdoctoral fellowship! https://vernadankers.com/blog/news-2025-12-video/ Thu, 18 Dec 2025 00:00:00 +0000 https://vernadankers.com/blog/news-2025-12-video/ Video available for "Memorisation: myth or mystery?", presented at the Ivado workshop on Deploying Autonomous Agents. Watch on YouTube. https://vernadankers.com/blog/news-2025-11-thesis/ Thu, 20 Nov 2025 00:00:00 +0000 https://vernadankers.com/blog/news-2025-11-thesis/ My PhD thesis is now available here. https://vernadankers.com/blog/news-2025-11-blackboxnlp/ Sat, 15 Nov 2025 00:00:00 +0000 https://vernadankers.com/blog/news-2025-11-blackboxnlp/ Attending BlackboxNLP and honored to give a keynote: "Memorization: myth or mystery?". Abstract: "In deep learning, the perspective on memorization of training examples is undergoing a paradigm shift. Previously linked to overfitting and poor generalization, memorization is now seen as benign, beneficial or concerning, depending on the data involved. This shift raises questions about the mystery that is memorization: what should and shouldn’t models memorize, how is memorization implemented internally, and, more fundamentally, can we talk about memorization as a phenomenon that is separate from generalization or is this a myth? https://vernadankers.com/blog/news-2025-08/ Fri, 01 Aug 2025 00:00:00 +0000 https://vernadankers.com/blog/news-2025-08/ Received the IJCAI 2025 JAIR award for Compositionality Decomposed! https://vernadankers.com/blog/news-2025-07/ Tue, 01 Jul 2025 00:00:00 +0000 https://vernadankers.com/blog/news-2025-07/ I defended my PhD and am joining Prof Siva Reddy as a postdoc at Mila and McGillNLP.