{"id":19,"date":"2023-06-29T17:42:13","date_gmt":"2023-06-29T17:42:13","guid":{"rendered":"https:\/\/projects.illc.uva.nl\/indeep\/?page_id=19"},"modified":"2026-10-19T19:03:14","modified_gmt":"2026-10-19T19:03:14","slug":"events","status":"publish","type":"page","link":"https:\/\/projects.illc.uva.nl\/indeep\/events\/","title":{"rendered":"Events"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\"><strong>What if&#8230; we want to maintain control\u00a0over AI?<\/strong><\/h2>\n\n\n\n<p class=\"has-text-align-left has-medium-font-size wp-block-paragraph\">On Friday 9 October, from 17:00 to 20:00, the UvA University Library hosted an evening on responsible artificial intelligence, organized by the National Research Agenda project \u2018InDeep\u2019. Beyond the hype and headlines, experts, artists, and visitors came together to explore the intellectual, ethical, and artistic dimensions of AI. A central question was how citizens can regain a sense of control over these technological developments. Visitors could choose from five thought-provoking sessions (two of them in Dutch), experience AI demos, and join the conversation about the future of AI. The evening concluded with drinks and light refreshments.<\/p>\n\n\n<style>.kb-image19_707afc-32.kb-image-is-ratio-size, .kb-image19_707afc-32 .kb-image-is-ratio-size{max-width:707px;width:100%;}.wp-block-kadence-column > .kt-inside-inner-col > .kb-image19_707afc-32.kb-image-is-ratio-size, .wp-block-kadence-column > .kt-inside-inner-col > .kb-image19_707afc-32 .kb-image-is-ratio-size{align-self:unset;}.kb-image19_707afc-32 figure{max-width:707px;}.kb-image19_707afc-32 .image-is-svg, .kb-image19_707afc-32 .image-is-svg img{width:100%;}.kb-image19_707afc-32 .kb-image-has-overlay:after{opacity:0.3;}<\/style>\n<div class=\"wp-block-kadence-image kb-image19_707afc-32\"><figure class=\"alignleft size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"280\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Entrepeneurs-3-1024x280.png\" alt=\"\" class=\"kb-img wp-image-1398\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Entrepeneurs-3-1024x280.png 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Entrepeneurs-3-300x82.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Entrepeneurs-3-768x210.png 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Entrepeneurs-3-1536x421.png 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Entrepeneurs-3.png 1822w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure><\/div>\n\n\n\n<p class=\"has-text-align-left has-medium-font-size wp-block-paragraph\"><strong>Entrepreneurs in action. Innovation and regulation in the Netherlands<\/strong>.<br>After the opening of the evening, a panel discussion (in English) took place with <strong>Julia Berezutskaya<\/strong> (UCANACT), <strong>Tijmen Altena <\/strong>(Global Campus), and the audience. The panel led by <strong>Jelle Zuidema <\/strong>(UvA) investigated what Dutch AI companies need to flourish, and whether the interests of companies and citizens are aligned. A key question was whether it is good or bad for Dutch citizens if AI companies have to comply with strict regulations.<\/p>\n\n\n<style>.kb-image19_ca0879-88.kb-image-is-ratio-size, .kb-image19_ca0879-88 .kb-image-is-ratio-size{max-width:616px;width:100%;}.wp-block-kadence-column > .kt-inside-inner-col > .kb-image19_ca0879-88.kb-image-is-ratio-size, .wp-block-kadence-column > .kt-inside-inner-col > .kb-image19_ca0879-88 .kb-image-is-ratio-size{align-self:unset;}.kb-image19_ca0879-88 figure{max-width:616px;}.kb-image19_ca0879-88 .image-is-svg, .kb-image19_ca0879-88 .image-is-svg img{width:100%;}.kb-image19_ca0879-88 .kb-image-has-overlay:after{opacity:0.3;}<\/style>\n<div class=\"wp-block-kadence-image kb-image19_ca0879-88\"><figure class=\"alignright size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"338\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/TEchtonic-1024x338.png\" alt=\"\" class=\"kb-img wp-image-1400\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/TEchtonic-1024x338.png 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/TEchtonic-300x99.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/TEchtonic-768x253.png 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/TEchtonic-1536x506.png 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/TEchtonic.png 1799w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure><\/div>\n\n\n\n<p class=\"has-text-align-left has-medium-font-size wp-block-paragraph\"><strong>The Techtonic Shift. Paul Verhagen on AI, Power, and Democracy<\/strong>.<br>In this theatre lecture (in English), <strong>Paul Verhagen <\/strong>(The Hague Centre for Strategic Studies) explored how the global AI race is fundamentally reshaping geopolitical power dynamics. He asked whether a world increasingly dominated by AI can remain democratic and liberal, and whether Europe can afford to stay technologically dependent on the United States and China. Verhagen examined the need to reform our societal institutions and presented insights from his new book, The Techtonic Shift.<\/p>\n\n\n<style>.kb-image19_e65283-2e.kb-image-is-ratio-size, .kb-image19_e65283-2e .kb-image-is-ratio-size{max-width:404px;width:100%;}.wp-block-kadence-column > .kt-inside-inner-col > .kb-image19_e65283-2e.kb-image-is-ratio-size, .wp-block-kadence-column > .kt-inside-inner-col > .kb-image19_e65283-2e .kb-image-is-ratio-size{align-self:unset;}.kb-image19_e65283-2e figure{max-width:404px;}.kb-image19_e65283-2e .image-is-svg, .kb-image19_e65283-2e .image-is-svg img{width:100%;}.kb-image19_e65283-2e .kb-image-has-overlay:after{opacity:0.3;}<\/style>\n<div class=\"wp-block-kadence-image kb-image19_e65283-2e\"><figure class=\"alignleft size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"524\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Creatievelingen2-1024x524.png\" alt=\"\" class=\"kb-img wp-image-1419\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Creatievelingen2-1024x524.png 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Creatievelingen2-300x154.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Creatievelingen2-768x393.png 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Creatievelingen2.png 1158w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure><\/div>\n\n\n\n<p class=\"has-text-align-left has-medium-font-size wp-block-paragraph\"><strong>Creatives Lose Out. What AI Is Doing to Creative Work<\/strong>.<br>In this theatre lecture (in Dutch), expert <strong>Jaap Jumelet<\/strong> (University of Groningen) explained why AI needs so much data, for example to generate music. He also addressed why composers, artists, and writers are so angry that AI companies used their work without permission, and why it is technically so difficult to trace whose work played a role in AI creations.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"alignright size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"338\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Auteursrechten3-1024x338.png\" alt=\"\" class=\"wp-image-1432\" style=\"aspect-ratio:3.029679626462022;width:544px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Auteursrechten3-1024x338.png 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Auteursrechten3-300x99.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Auteursrechten3-768x253.png 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Auteursrechten3-1536x506.png 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Auteursrechten3.png 1799w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-text-align-left has-medium-font-size wp-block-paragraph\"><strong>Copyright Turned Upside Down. Discover the future of copyright and AI<\/strong>.<br>In this discussion (in Dutch), moderated by <strong>Iris Hendrickx<\/strong> (Radboud University), <strong>Daan Archer<\/strong> (CEO and co-founder of Copyright Delta) and <strong>Kees Bulder <\/strong>(legal counsel at Buma\/Stemra) addressed the accusations that AI companies have infringed copyright on a massive scale, and whether fair AI is possible. What does copyright still mean in a time when so much text, music, and imagery is generated by AI?<\/p>\n\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><strong>AI and the Builders. Meet the builders of tomorrow\u2019s AI: live interview and discussion<\/strong>.<br>In a live interview and discussion (in English) with <strong>Onno Zoeter<\/strong> (head of AI research at Booking) and <strong>Akash Raj Komarlu <\/strong>(CTO and co-founder of Whispp), we asked what it takes to be successful in the new AI-dominated society. What skills do workers need? How can citizens remain safe and prosperous? Is it still fun to work in AI, or is everyone stressed out by the rapid progress and new perspectives on its downsides? What new shocks should we prepare for? The audience heard it from the builders who are building next year\u2019s AI today.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"214\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Builders-1024x214.png\" alt=\"\" class=\"wp-image-1426\" style=\"aspect-ratio:4.785455915599096;width:982px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Builders-1024x214.png 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Builders-300x63.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Builders-768x160.png 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Builders-1536x320.png 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/Builders-2048x427.png 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\">During the evening, in collaboration with the <em>Amsterdam Data Science Centre<\/em>, visitors could explore a range of interactive AI demos, including fake speech detection, emotion recognition, and a game that pitted human creativity against AI perception. They could also browse a curated selection of books on AI, innovation, and the future of technology. After the last session, drinks and light bites were served, giving participants the chance to continue the conversation and network with one another.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"777\" height=\"237\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/demos.png\" alt=\"\" class=\"wp-image-1435\" style=\"aspect-ratio:3.2786312388738708;width:938px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/demos.png 777w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/demos-300x92.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/10\/demos-768x234.png 768w\" sizes=\"auto, (max-width: 777px) 100vw, 777px\" \/><\/figure>\n\n\n<style>.wp-block-kadence-spacer.kt-block-spacer-19_a6bbb6-6e .kt-block-spacer{height:60px;}.wp-block-kadence-spacer.kt-block-spacer-19_a6bbb6-6e .kt-divider{border-top-width:3px;height:1px;border-top-color:rgba(47, 90, 174, 0.60);width:80%;border-top-style:solid;}<\/style>\n<div class=\"wp-block-kadence-spacer aligncenter kt-block-spacer-19_a6bbb6-6e\"><div class=\"kt-block-spacer kt-block-spacer-halign-center\"><hr class=\"kt-divider\"\/><\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Playing with Speech Technology \u2013 Weekend of Science<\/strong><\/h2>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"205\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/Scherm\u00adafbeelding-2026-09-22-om-10.09.09-1024x205.png\" alt=\"\" class=\"wp-image-1286\" style=\"aspect-ratio:4.995326719685169;width:520px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/Scherm\u00adafbeelding-2026-09-22-om-10.09.09-1024x205.png 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/Scherm\u00adafbeelding-2026-09-22-om-10.09.09-300x60.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/Scherm\u00adafbeelding-2026-09-22-om-10.09.09-768x154.png 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/Scherm\u00adafbeelding-2026-09-22-om-10.09.09-1536x307.png 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/Scherm\u00adafbeelding-2026-09-22-om-10.09.09.png 1720w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><br>On Sunday 4 October, from 14:00 to 17:00, researchers from the InDeep project explored the frontiers of AI and speech technology at Spoorzone Tilburg during the \u2018Weekend van de Wetenschap\u2019. Through live demonstrations, visitors could interact directly with speech systems, probe their capabilities, and see firsthand where today\u2019s technology excels, and where it still falls short. One of the questions they tried to answer: can a smart computer guess your emotions from the sound of your voice? <\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"alignleft size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"636\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/977abbf3-ea9d-f2ad-c6b6-5630fbf47ba0-1024x636.jpg\" alt=\"\" class=\"wp-image-1292\" style=\"aspect-ratio:1.6100965355844372;width:282px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/977abbf3-ea9d-f2ad-c6b6-5630fbf47ba0-1024x636.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/977abbf3-ea9d-f2ad-c6b6-5630fbf47ba0-300x186.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/977abbf3-ea9d-f2ad-c6b6-5630fbf47ba0-768x477.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/977abbf3-ea9d-f2ad-c6b6-5630fbf47ba0-1536x954.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/09\/977abbf3-ea9d-f2ad-c6b6-5630fbf47ba0.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\">The event was free and open to everyone, of all ages, curious about AI \u2014 no technical background required. InDeep brings together researchers from Tilburg University and partner institutions working on how machines learn to listen, speak, and understand language and music. <\/p>\n\n\n<style>.wp-block-kadence-spacer.kt-block-spacer-19_5d5224-59 .kt-block-spacer{height:60px;}.wp-block-kadence-spacer.kt-block-spacer-19_5d5224-59 .kt-divider{border-top-width:3px;height:1px;border-top-color:rgba(47, 90, 174, 0.60);width:80%;border-top-style:solid;}<\/style>\n<div class=\"wp-block-kadence-spacer aligncenter kt-block-spacer-19_5d5224-59\"><div class=\"kt-block-spacer kt-block-spacer-halign-center\"><hr class=\"kt-divider\"\/><\/div><\/div>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<h2 class=\"wp-block-heading\">&nbsp;<strong>InDeep Cum Laude Doctorate and Consortium Workshop<\/strong><\/h2>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"692\" height=\"510\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/IMG_5488-3.png\" alt=\"\" class=\"wp-image-1148\" style=\"width:454px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/IMG_5488-3.png 692w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/IMG_5488-3-300x221.png 300w\" sizes=\"auto, (max-width: 692px) 100vw, 692px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\">Our InDeep PhD candidate&nbsp;<strong>Gabriele Sarti<\/strong>&nbsp;successfully defended his dissertation,&nbsp;<em>From Insights to Impact: Actionable Interpretability for Neural Machine Translation<\/em>, on 11 December 2025 at the&nbsp;University of Groningen. Gabriele is the&nbsp;first PhD candidate within InDeep to complete and defend his doctorate, and he did so with outstanding expertise and enthusiasm, earning the distinction of&nbsp;<strong>cum laude<\/strong>. We warmly congratulate <strong>Dr. Gabriele Sarti <\/strong>and wish him every success in his future career. <br><span style=\"caret-color: rgb(0, 0, 0); font-family: var(--bodyfontfamily);\">You can read his dissertation <\/span><a style=\"caret-color: rgb(0, 0, 0); font-family: var(--bodyfontfamily);\" href=\"https:\/\/research.rug.nl\/en\/publications\/from-insights-to-impact-actionable-interpretability-for-neural-ma\/\">here<\/a><span style=\"caret-color: rgb(0, 0, 0); font-family: var(--bodyfontfamily);\">.<\/span><\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"764\" height=\"238\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/IMG_4374_2.png\" alt=\"\" class=\"wp-image-1147\" style=\"width:888px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/IMG_4374_2.png 764w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/IMG_4374_2-300x93.png 300w\" sizes=\"auto, (max-width: 764px) 100vw, 764px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\">To mark this occasion, we hosted an&nbsp;<strong>InDeep Consortium Workshop<\/strong>&nbsp;prior to the defense. The workshop featured presentations by <em>Gabriele<\/em>, project leader&nbsp;<em>Jelle Zuidema<\/em>, as well as invited talks by&nbsp;<em>Barbara Plank<\/em>&nbsp;(LMU Munich),&nbsp;<em>Ivan Titov<\/em>&nbsp;(University of Amsterdam\/University of Edinburgh), and&nbsp;<em>Eva Vanmassenhove<\/em>&nbsp;(Tilburg University). Following the defense, the celebration continued with a&nbsp;poster session and drinks.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"168\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/Groningen_tg-1024x168.png\" alt=\"\" class=\"wp-image-1149\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/Groningen_tg-1024x168.png 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/Groningen_tg-300x49.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/Groningen_tg-768x126.png 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/Groningen_tg.png 1329w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><strong>Eva Vanmassenhove&nbsp;<\/strong>&nbsp;<em>Losing our tail, again: On (Un)natural Selection and Language Technologies<\/em><br><strong>Ivan Titov&nbsp;<\/strong><em>Why Do Interpretability Research? From Frustration to Actionability<br><\/em><strong>Gabriele Sarti&nbsp;<\/strong><em>From Insight to Impact: Actionable Interpretability for Neural Machine&nbsp;<\/em><br><strong>Barbara Plank&nbsp;<\/strong><em>From Arithmetic Validation Gaps to Explanation Variation<\/em><br><strong>Willem Zuidema<\/strong><em> Mechanistic Interpretability of Time Series Transformers (and its Relevance for Language and Reasoning)<\/em><\/p>\n\n\n<style>.wp-block-kadence-spacer.kt-block-spacer-19_2194bf-8b .kt-block-spacer{height:60px;}.wp-block-kadence-spacer.kt-block-spacer-19_2194bf-8b .kt-divider{border-top-width:3px;height:1px;border-top-color:rgba(47, 90, 174, 0.60);width:80%;border-top-style:solid;}<\/style>\n<div class=\"wp-block-kadence-spacer aligncenter kt-block-spacer-19_2194bf-8b\"><div class=\"kt-block-spacer kt-block-spacer-halign-center\"><hr class=\"kt-divider\"\/><\/div><\/div>\n\n\n\n<p class=\"has-large-font-size wp-block-paragraph\"><strong>InDeep Masterclass at Deloitte<\/strong><\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"alignright size-full is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"576\" height=\"549\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/Deloitte22.png\" alt=\"\" class=\"wp-image-1172\" style=\"width:512px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/Deloitte22.png 576w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2026\/01\/Deloitte22-300x286.png 300w\" sizes=\"auto, (max-width: 576px) 100vw, 576px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\">In December 2025 InDeep presented their <strong>Interpreting and Understanding LLMs and Other Deep Learning Models<\/strong> Masterclass at Deloitte Amsterdam.&nbsp;This masterclass offered an in-depth exploration of explainability and interpretability in large language models and other deep learning architectures. Featuring a series of expert-led presentations and an interactive hands-on tutorial, the event combined theoretical perspectives with practical methods for understanding how modern AI systems make decisions. Speakers addressed key challenges in explainable AI, including the trade-offs between faithfulness and comprehensibility of explanations, the limits of existing XAI techniques, and emerging approaches such as mechanistic interpretability, attribution methods, and behavioral evaluation through challenge sets. By bringing together academic expertise and applied discussion, the masterclass equipped participants with a deeper understanding of the inner workings, limitations, and evaluation of generative AI systems.<br><br><strong>Speakers:<\/strong><br>\u2013 Gabriele Sarti (RUG Groningen)&nbsp;\u2013&nbsp;<em>Interpreting Large Language Models<\/em><br>\u2013 Jelle Zuidema (UvA) \u2013&nbsp;<em>Too Much Information: How do we generate comprehensible and Faithful explanations when the input or output is\u2026just a bit much?<\/em><br>\u2013 Antske Fokkens (VU) \u2013&nbsp;<em>Explanatory Evaluation and Robustness<\/em><\/p>\n\n\n<style>.wp-block-kadence-spacer.kt-block-spacer-19_4f0e00-9d .kt-block-spacer{height:60px;}.wp-block-kadence-spacer.kt-block-spacer-19_4f0e00-9d .kt-divider{border-top-width:3px;height:1px;border-top-color:rgba(47, 90, 174, 0.60);width:80%;border-top-style:solid;}<\/style>\n<div class=\"wp-block-kadence-spacer aligncenter kt-block-spacer-19_4f0e00-9d\"><div class=\"kt-block-spacer kt-block-spacer-halign-center\"><hr class=\"kt-divider\"\/><\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">October Workshop: How Computing is Changing the World<\/h2>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"360\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/09\/QISS_Indeep-1024x360.png\" alt=\"\" class=\"wp-image-984\" style=\"width:566px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/09\/QISS_Indeep-1024x360.png 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/09\/QISS_Indeep-300x105.png 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/09\/QISS_Indeep-768x270.png 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/09\/QISS_Indeep-1536x540.png 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/09\/QISS_Indeep.png 1605w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><strong>Workshop: &#8220;How Computing is Changing the World: Exploring Synergies and Challenges of AI and Quantum Technologies for Society&#8221;<\/strong><\/p>\n\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\">In October 2025, the&nbsp;<a href=\"https:\/\/projects.illc.uva.nl\/quantumimpact\/\">QISS<\/a>&nbsp; and InDeep research groups met at the Sustainalab (Matrix One) in Amsterdam to explore in a joint workshop the synergies and challenges of AI and quantum technologies. As both fields advance rapidly, the workshop focused on their technological progress as well as their ethical, legal, and societal implications.<\/p>\n\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\">The morning session addressed technological perspectives. After a welcome, InDeep project leader <strong>Jelle Zuidema<\/strong> outlined the current state of AI, highlighting the black-box problem. <strong>Mehrnoosh Sadrzadeh<\/strong> (UCL) and <strong>Martha Lewis<\/strong> (UvA) followed with quantum-inspired approaches to modelling language and vision in AI. <strong>Ronald de Wolf<\/strong> (CWI &amp; UvA, QuSoft) concluded with an overview of quantum computing, linking it to Zuidema\u2019s AI analysis. A panel discussion with <strong>Christian Schaffner<\/strong> (UvA, QuSoft) and chaired by <strong>Joris van Hoboken<\/strong> (UvA) explored common challenges, research stages, and opportunities across the two domains.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"287\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_3982-s-1024x287.jpg\" alt=\"\" class=\"wp-image-1096\" style=\"width:954px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_3982-s-1024x287.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_3982-s-300x84.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_3982-s-768x215.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_3982-s-1536x430.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_3982-s-2048x573.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><br>The afternoon session focused on societal themes. <strong>Evert van Nieuwenburg<\/strong> (Leiden University) opened with a presentation on how quantum games, such as Quantum Tic-Tac-Toe, can help people learn quantum principles. <strong>Eline de Jong<\/strong> (UvA) shared five lessons for embedding new technologies in society, suggesting they apply to quantum tech as well. <strong>Angela van Sprang <\/strong>(UvA, ILLC) presented concept-bottleneck models as tools for improved AI oversight, and <strong>Matteo Fabbri<\/strong> (UvA) argued for proactive standards for recommender systems and other emerging technologies. In the panel discussion, led by Zuidema, participants from both AI and quantum backgrounds reflected on responsible innovation and the risks and opportunities of uncertain technological futures.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"295\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_samen-s-1024x295.jpg\" alt=\"\" class=\"wp-image-1098\" style=\"width:709px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_samen-s-1024x295.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_samen-s-300x86.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_samen-s-768x221.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_samen-s-1536x442.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/12\/IMG_samen-s-2048x590.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><br>The workshop offered an engaging exchange at the intersection of technology, ethics, and society. Participants valued the lively discussions and expressed enthusiasm for organising a follow-up event in 2026, potentially establishing a recurring series on the intersections of AI and quantum technologies.<\/p>\n\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><em>The&nbsp;Quantum Impact on Societal Security&nbsp;(QISS) consortium is based at the University of Amsterdam and analyses the ethical, legal and societal impact of the upcoming society-wide transition to quantum-safe cryptography. The consortium&#8217;s objective is to contribute to the creation of a Dutch ecosystem where quantum-safe cryptography can thrive, and mobilize this ecosystem to align technological applications with ethical, legal, and social values.&nbsp;<br><\/em><br><em>This workshop is made possible with funding from the Quantum Delta Netherlands growthfund program.&nbsp;<\/em><\/p>\n\n\n<style>.wp-block-kadence-spacer.kt-block-spacer-19_a831c3-77 .kt-block-spacer{height:60px;}.wp-block-kadence-spacer.kt-block-spacer-19_a831c3-77 .kt-divider{border-top-width:3px;height:1px;border-top-color:rgba(47, 90, 174, 0.60);width:80%;border-top-style:solid;}<\/style>\n<div class=\"wp-block-kadence-spacer aligncenter kt-block-spacer-19_a831c3-77\"><div class=\"kt-block-spacer kt-block-spacer-halign-center\"><hr class=\"kt-divider\"\/><\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>InDeep at Interspeech<\/strong> 2025<\/h2>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized has-custom-border\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"371\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/07\/Interspeech_indeep-1024x371.jpg\" alt=\"\" class=\"wp-image-876\" style=\"border-width:3px;border-radius:25px;width:651px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/07\/Interspeech_indeep-1024x371.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/07\/Interspeech_indeep-300x109.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/07\/Interspeech_indeep-768x278.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/07\/Interspeech_indeep-1536x556.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2025\/07\/Interspeech_indeep.jpg 1822w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><br>Many InDeep researchers presented at the&nbsp;<a href=\"https:\/\/www.interspeech2025.org\/home\" target=\"_blank\" rel=\"noreferrer noopener\">Interspeech<\/a> 2025 conference in Rotterdam, one of the largest and most prominent conferences on speech and speech technology world-wide. Therefore, we also organized a<strong>&nbsp;social event<\/strong>&nbsp;including drinks and snacks.&nbsp;<\/p>\n\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><strong>This was InDeep at Interspeech 2025<\/strong>:<br><strong>Sunday 17 August<\/strong><br><strong>&#8211;&nbsp;<\/strong><a href=\"https:\/\/www.interspeech2025.org\/science-fest\" target=\"_blank\" rel=\"noreferrer noopener\"><strong>Speech Science Festival&nbsp;<\/strong><\/a><strong>10.00-17.30.&nbsp;<\/strong>InDeep provided a demonstration of emotion-decoding and removal. There were many other demonstrations for a broad audience. <br>&#8211;&nbsp;<a href=\"https:\/\/www.interspeech2025.org\/tutorials\" target=\"_blank\" rel=\"noreferrer noopener\"><strong>Tutorial<\/strong><\/a><strong>, 15.30-18.30<\/strong>:&nbsp;InDeep members organized the tutorial&nbsp;<em>Interpretability Techniques for Speech Model<\/em>.&nbsp;<br>&#8211;&nbsp;<strong>InDeep Social, 19.00:<\/strong>&nbsp;InDeep held a social event with drinks and snacks in the Stuurboord bar at the&nbsp;<a href=\"https:\/\/maps.app.goo.gl\/uCEyCyfpe7ji59S39\" target=\"_blank\" rel=\"noreferrer noopener\">Foodhallen<\/a>. <br><br><strong>Monday 18 August<br>&#8211;&nbsp;<\/strong><a href=\"https:\/\/www.interspeech2025.org\/programme-full\" target=\"_blank\" rel=\"noreferrer noopener\"><strong>Special Session<\/strong><\/a>&nbsp;on&nbsp;<em>Interpretability in Audio and Speech&nbsp;<\/em>Technology, 11:00-13:00.&nbsp;Session with three papers by InDeep authors:&nbsp;<br>+ <em>On the reliability of feature attribution methods for speech classification.<\/em>&nbsp;<em>&nbsp;<\/em><br>+ <em>Word stress in self-supervised speech models: A cross-linguistic comparison.<\/em><br>+ <em>What do self-supervised speech models know about Dutch? Analyzing advantages of language-specific pre-training.<\/em><\/p>\n\n\n<style>.wp-block-kadence-spacer.kt-block-spacer-19_232115-bb .kt-block-spacer{height:60px;}.wp-block-kadence-spacer.kt-block-spacer-19_232115-bb .kt-divider{border-top-width:3px;height:1px;border-top-color:rgba(47, 90, 174, 0.60);width:80%;border-top-style:solid;}<\/style>\n<div class=\"wp-block-kadence-spacer aligncenter kt-block-spacer-19_232115-bb\"><div class=\"kt-block-spacer kt-block-spacer-halign-center\"><hr class=\"kt-divider\"\/><\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">Consortium Meeting Groningen<\/h2>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized has-custom-border\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"561\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/IMG_5903-bewerkt2-kopie-1024x561.jpg\" alt=\"\" class=\"wp-image-717\" style=\"border-width:3px;border-radius:25px;width:663px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/IMG_5903-bewerkt2-kopie-1024x561.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/IMG_5903-bewerkt2-kopie-300x164.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/IMG_5903-bewerkt2-kopie-768x421.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/IMG_5903-bewerkt2-kopie-1536x842.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/IMG_5903-bewerkt2-kopie-2048x1123.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><br>On&nbsp;<strong>Friday 20 September 2024<\/strong> InDeep met&nbsp;in the&nbsp;<strong>Groningen House of Connections&nbsp;<\/strong>for the next Consortium Meeting.&nbsp;In the morning RUG researchers&nbsp;<em>Daniel Herrman&nbsp;<\/em>(Philosophy of AI) and&nbsp;<em>Yevgen Matusevych<\/em>&nbsp;(Cognitive Plausibility of Modern LMs) gave illuminating talks on standards for belief representations in LLMs and bias in visually grounded speech models.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized has-custom-border\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"356\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/presentaties-s-1024x356.jpg\" alt=\"\" class=\"wp-image-718\" style=\"border-width:3px;border-radius:25px;width:731px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/presentaties-s-1024x356.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/presentaties-s-300x104.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/presentaties-s-768x267.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/presentaties-s-1536x534.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/presentaties-s-2048x712.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><br>In the afternoon highlights from the various InDeep academic partners were shared, there was a poster session by researchers on their latest research results, as well as fruitful discussions on the future of interpretability methods in language, speech and music models and future directions of the InDeep project.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large has-custom-border\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"242\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/posters-s2-1024x242.jpg\" alt=\"\" class=\"wp-image-715\" style=\"border-width:3px;border-radius:25px\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/posters-s2-1024x242.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/posters-s2-300x71.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/posters-s2-768x182.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/posters-s2-1536x364.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/11\/posters-s2-2048x485.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n<style>.wp-block-kadence-spacer.kt-block-spacer-19_573c29-ca .kt-block-spacer{height:60px;}.wp-block-kadence-spacer.kt-block-spacer-19_573c29-ca .kt-divider{border-top-width:3px;height:1px;border-top-color:rgba(47, 90, 174, 0.60);width:80%;border-top-style:solid;}<\/style>\n<div class=\"wp-block-kadence-spacer aligncenter kt-block-spacer-19_573c29-ca\"><div class=\"kt-block-spacer kt-block-spacer-halign-center\"><hr class=\"kt-divider\"\/><\/div><\/div>\n\n\n\n<h3 class=\"wp-block-heading\">Consortium Meeting Amsterdam<\/h3>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"576\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/02\/InDeep-Consortium_03-24_final_2s-1024x576.jpg\" alt=\"\" class=\"wp-image-623\" style=\"width:715px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/02\/InDeep-Consortium_03-24_final_2s-1024x576.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/02\/InDeep-Consortium_03-24_final_2s-300x169.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/02\/InDeep-Consortium_03-24_final_2s-768x432.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/02\/InDeep-Consortium_03-24_final_2s-1536x864.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/02\/InDeep-Consortium_03-24_final_2s-2048x1152.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><br>On <strong>Friday 8 March <\/strong><strong>2024<\/strong> InDeep had a Consortium meeting in the <strong>Amsterdam University Library<\/strong>.&nbsp;It was a day with both talks on recent progress in research, a tutorial, and lots of opportunity to discuss the strengths and weaknesses of interpretability techniques in text, translation, speech and music.&nbsp;<\/p>\n\n\n<style>.wp-block-kadence-spacer.kt-block-spacer-19_5a981b-83 .kt-block-spacer{height:60px;}.wp-block-kadence-spacer.kt-block-spacer-19_5a981b-83 .kt-divider{border-top-width:3px;height:1px;border-top-color:rgba(47, 90, 174, 0.60);width:80%;border-top-style:solid;}<\/style>\n<div class=\"wp-block-kadence-spacer aligncenter kt-block-spacer-19_5a981b-83\"><div class=\"kt-block-spacer kt-block-spacer-halign-center\"><hr class=\"kt-divider\"\/><\/div><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">InDeep Masterclass: Explaining Foundation Models<\/h2>\n\n\n\n<figure class=\"wp-block-image size-large is-resized has-custom-border\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"239\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/fronttriple-2-1024x239.jpg\" alt=\"\" class=\"wp-image-475\" style=\"border-width:3px;border-radius:25px;width:969px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/fronttriple-2-1024x239.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/fronttriple-2-300x70.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/fronttriple-2-768x180.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/fronttriple-2-1536x359.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/fronttriple-2-2048x479.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><\/p>\n\n\n<div class=\"wp-block-image is-style-default\">\n<figure class=\"alignright size-large is-resized has-custom-border\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"828\" src=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/Save-the-date_new_extraAAI_3_omlijnt2-2-1024x828.jpg\" alt=\"\" class=\"wp-image-473\" style=\"border-width:3px;border-radius:25px;width:573px;height:auto\" srcset=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/Save-the-date_new_extraAAI_3_omlijnt2-2-1024x828.jpg 1024w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/Save-the-date_new_extraAAI_3_omlijnt2-2-300x243.jpg 300w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/Save-the-date_new_extraAAI_3_omlijnt2-2-768x621.jpg 768w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/Save-the-date_new_extraAAI_3_omlijnt2-2-1536x1243.jpg 1536w, https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2023\/12\/Save-the-date_new_extraAAI_3_omlijnt2-2-2048x1657.jpg 2048w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"has-text-align-left has-medium-font-size wp-block-paragraph\">On Thursday 2 November 2023, <strong>InDeep<\/strong> organized a <strong>Masterclass<\/strong> on&nbsp;<em>Explaining Foundation Models<\/em>.&nbsp;The event featured&nbsp;some inspiring talks, hands-on experience with explainability techniques for Large Language Models, and the opportunity to share insights and experiences.&nbsp;After the Masterclass there was a Meetup organized by Indeep and Amsterdam Al, on&nbsp;<em>Alternatives for ChatGPT: How good are open source and in-house LLMs?<\/em>&nbsp;To learn more about this event, you can download <a href=\"https:\/\/projects.illc.uva.nl\/indeep\/wp-content\/uploads\/2024\/01\/InDeep_Masterclass_Meetup_.pdf\">here<\/a> an overview of the presentations and slides.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"has-medium-font-size wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n","protected":false},"excerpt":{"rendered":"<p>What if&#8230; we want to maintain control\u00a0over AI? On Friday 9 October, from 17:00 to 20:00, the UvA University Library hosted an evening on responsible artificial intelligence, organized by the National Research Agenda project \u2018InDeep\u2019. Beyond the hype and headlines, experts, artists, and visitors came together to explore the intellectual, ethical, and artistic dimensions of&hellip;&nbsp;<a href=\"https:\/\/projects.illc.uva.nl\/indeep\/events\/\" rel=\"bookmark\">Read More &raquo;<span class=\"screen-reader-text\">Events<\/span><\/a><\/p>\n","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"neve_meta_sidebar":"","neve_meta_container":"","neve_meta_enable_content_width":"","neve_meta_content_width":0,"neve_meta_title_alignment":"","neve_meta_author_avatar":"","neve_post_elements_order":"","neve_meta_disable_header":"","neve_meta_disable_footer":"","neve_meta_disable_title":"","footnotes":""},"class_list":["post-19","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/projects.illc.uva.nl\/indeep\/wp-json\/wp\/v2\/pages\/19","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/projects.illc.uva.nl\/indeep\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/projects.illc.uva.nl\/indeep\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/projects.illc.uva.nl\/indeep\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/projects.illc.uva.nl\/indeep\/wp-json\/wp\/v2\/comments?post=19"}],"version-history":[{"count":178,"href":"https:\/\/projects.illc.uva.nl\/indeep\/wp-json\/wp\/v2\/pages\/19\/revisions"}],"predecessor-version":[{"id":1437,"href":"https:\/\/projects.illc.uva.nl\/indeep\/wp-json\/wp\/v2\/pages\/19\/revisions\/1437"}],"wp:attachment":[{"href":"https:\/\/projects.illc.uva.nl\/indeep\/wp-json\/wp\/v2\/media?parent=19"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}