Kristina Fort’s Post

In #Berlin attending the 𝐆𝐥𝐨𝐛𝐚𝐥 𝐓𝐞𝐜𝐡 𝐌𝐢𝐧𝐢𝐧𝐠 𝐂𝐨𝐧𝐟𝐞𝐫𝐞𝐧𝐜𝐞 and the 28𝐭𝐡 𝐈𝐧𝐭𝐞𝐫𝐧𝐚𝐭𝐢𝐨𝐧𝐚𝐥 𝐂𝐨𝐧𝐟𝐞𝐫𝐞𝐧𝐜𝐞 𝐨𝐧 𝐒𝐜𝐢𝐞𝐧𝐜𝐞, 𝐓𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐲 𝐚𝐧𝐝 𝐈𝐧𝐧𝐨𝐯𝐚𝐭𝐢𝐨𝐧 𝐈𝐧𝐝𝐢𝐜𝐚𝐭𝐨𝐫𝐬 (𝐒𝐓𝐈 2024) for the days to come! Looking at this year’s program, the focus is on topics such as the impact of AI, emerging technologies, and machine learning on scientific research and societal developments. At the STI 2024 conference, themes include the societal impact of science, geographical effects on collaboration, research quality, and interdisciplinarity. There is a strong emphasis on the influence of open science, bibliometric methods, and innovation policies. The GTM 2024 conference highlights advanced analytical techniques, innovative approaches in tech mining, and the use of large language models in research. Both conferences address critical issues like academic mobility, diversity in research, and the ethical implications of AI in research and bibliometrics, reflecting a commitment to understanding and improving the intersection of science, technology, and society. Also, a shout-out to our session today at 14:00 at Lux Room on "𝐀𝐧𝐚𝐥𝐲𝐳𝐢𝐧𝐠 𝐓𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐢𝐜𝐚𝐥 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐬 𝐚𝐧𝐝 𝐄𝐜𝐨𝐬𝐲𝐬𝐭𝐞𝐦𝐬", where we discuss the nuances of methods to analyze ecosystems of science, technology, and innovation.

What if you could unlock the secrets of human behavior using cutting-edge technology? 🧠✨ Today, we’re diving deep into the power of Generative AI in social science research methods at an inspiring symposium in Melbourne. We're in the company of some of the sharpest minds in the field, collaborating to push the boundaries of what’s possible. Imagine the insights we could gain, the patterns we could identify, and the solutions we could create by harnessing the potential of AI. Curious about how Generative AI can revolutionize social science? Join the conversation and explore with us! 🔗 [Discover more in our full discussion](https://lnkd.in/gpxjqNfp) #GenerativeAI #SocialScience #Innovation

Last week, I visited the The London School of Economics and Political Science (LSE) for the first time in many years. As an alumnus, the campus felt familiar yet foreign. Familiar with the same buzz of students and energy of intellectual discussions, but foreign in the sense that that school has massively expanded, with numerous new buildings and facilities that were not previously there just 4-5 years ago! One of the reasons for my visit (I shall share the other in another post) was for the panel discussion, "Developing AI Literacy in the Classroom and Beyond". Expertly moderated by Professor Emma McCoy, and enriched with the perspectives of four diverse panelists, Dr. Jillian Terry, Cecilia N., Cody Dodd, and Ashley Winton, the event offered profound insights into the evolving landscape of #AI, #education, and #policy. Here are some of my key takeaways: 1️⃣ Rise of Prompt Engineering: Just as a solid understanding of Boolean search was previously critical for academic exploration, #promptengineering is becoming an essential skill for leveraging AI effectively, not just for students, but more widely for the general population. 2️⃣ Navigating AI Regulation: The panel explored the contrasting regulatory stances of different countries (e.g., the more deterministic stance adopted by Canada vs. the more laissez faire approach taken by the UK). More broadly, the rapid evolution of AI regulation underscores the need for agility when navigating this space. 3️⃣ Championing Responsible AI: Some panelists advocated for embedding #responsibleai practices from inception, highlighting it as a strategic advantage rather than a compliance exercise. This approach not only safeguards reputations but also positions organisations as leaders in ethical AI deployment. 4️⃣ Adaptive AI Policies: The rapidly changing technological landscape and the emergence of new, sometimes unforeseen, use cases requires organisations to adopt a dynamic approach toward AI policies and guardrails. 5️⃣ Evolution of Education: A significant discussion point was the role of AI in education, particularly its impact on teaching methods. While some institutions seek a return to pen and paper to preserve traditional standards, the panelists agreed that such measures are impractical in our technology-driven world. Instead, the focus should shift towards developing skills in analysis, real-world application, and ethical usage of AI tools. 6️⃣ Implications for AI Access: Post the event, I had an interesting discussion on the dual-edged nature of AI access, including balancing the potential for AI to exacerbate disparities alongside its ability to democratise access to powerful capabilities. The importance of fostering AI literacy was underscored as a critical endeavor for equitable global development. My thanks to the organising team for an excellent discussion, as well as Ceri Jones, Tshokey Mindara, and Shontae Romain from the LSE Alumni Engagement team for alerting me to the event!

New Post: New project within the CRC 1646 “Linguistic Creativity” to start in April 2024 A new Collaborative Research Center (CRC) on “Linguistic Creativity in Communication” funded by DFG is about to start in 2024. We are part of it with a project on multimodal creativity in speech-gesture production. In collaboration with Joana Cholin (Psycholinguistics) we will investigate how humans and AI models can use gesture to accompany newly created linguistic constructions, when given communicative resources are not available or not adequate. - https://lnkd.in/edMaFhE8

Super excited, the results of deepfake experiment we conducted got accepted to #IC2S2 2024! 10th International Conference on Computational Social Science Philadelphia, USA | July 17-20, 2024 https://ic2s2-2024.org/ J. Chen, D. Gamage, M. Haruyama, T. Igarashi, and K. Sasahara, Does hands-on learning experience of deepfakes reduce the intention to share them? We explored "Knowing how to make deepfakes, would you share in your social media?" .. tune in more for our study .. thanks for the team Sasahara Lab 🙏

🌍🚀 21 top students from all around the globe are now in Sofia for this year’s INSAIT Summer Research Fellowship Program 2024! Amongst them are international olympiad medalists, startup founders, gen AI hackathon winners and others. The research topics they work on at INSAIT span large language models, 3D understanding, safety of generative AI, security of machine learning, quantum computation, and more. 🔗 Learn more about these amazing students and their projects in the article - link in comments.

Excited to share the results of some recent research we've been working on as part of the Supervised Program for Alignment Research. Our team's been diving into the inner workings of LLMs - with the eventual goal of making them safer and more secure. We tackled the question: Do LLMs pre-plan their content? Using mechanistic interpretability techniques, we investigated how AI models plan ahead while generating paragraphs. Really appreciative of the chance to work with Nicky P., Lucile Ter-Minassian, Angelo Benoit and Lovkush Agarwal 🔸. For those interested in the detailed findings, our full paper's up on arXiv https://lnkd.in/dajwtv_6 and we've also got a version on Medium https://lnkd.in/eE6mbKNF. I'm hoping to continue this exploration at the intersection of AI and security! 😊

We are excited to congratulate our 2024 summer cohort for their outstanding research and impressive presentations at the annual BLAST Summer Symposium. This remarkable achievement highlights their dedication and hard work, and we are incredibly proud to celebrate their success. If you’re curious, feel free to view our symposium here: https://lnkd.in/gDQ2Sq_U The titles of the work presented are listed below: - Advancing Super-Resolution in Neural Radiance Fields via Variational Diffusion Strategies - Enhancing Lie Detection Accuracy: A Comparative Study of Classic ML, CNN, and GCN Models using Audio-Visual Features - Streaming Pitch Estimation Using Emformers - Enhancing LLM Conversational Acuity Using Pragmatics Measures

#GenAI #EnglishTeachers #ELT #CPD Missed the live webinar Evelina Galaczi and I presented or need to revise something about it? 🎥The recording is available! Let's keep the positive interaction we had during the live, like 👍 comment 💬 & share with your community of #teachers. Cambridge University Press & Assessment

Pursuing my PhD is a rewarding, new chapter and a great way to grow, while adding value to others. It is exciting to have our team’s work exploring genAI accepted at Society for Marketing Advances this fall! Conference theme: "The Age of AI." Very thankful for awesome co-authors and peers from my doctoral program (shoutout Mark Stroud and Logan Barrett!) at University of South Alabama Mitchell College of Business!