Wat we doen
Information drives the planet. We organize talks around implementations of information retrieval, in search engines, in recommender systems, or in conversational assistants. Our meetups are usually held on the last Friday of the month, at Science Park Amsterdam. Usually, we have two talks in a row, one industrial, the other academic, 25+5 minutes each, no marketing, just algorithms, followed by drinks. We also host ad hoc "single shot" events whenever an interesting visitor stops and shares their work.
Search Engines Amsterdam is supported by the ELLIS unit Amsterdam.
Follow @irlab_amsterdam on Twitter for the latest updates.
Aankomende evenementen
12
SEA Talks: January - Complex Deep Search
Amsterdam Science Park 904, Science Park 904, Amsterdam, NLHappy New Year! The SEA (Search Engines Amsterdam) Talk is coming 🎉!
We are excited to kick off 2026 with our first edition on January 30, focusing on Complex Deep Search and featuring two exciting speakers, Dan Li from Elsevier and Roxana Petcu from the University of Amsterdam.
This will be a hybrid event, taking place both in person and online.
Zoom link: https://uva-live.zoom.us/j/65011610507
Location: Science Park 904, Room C3.161
Speaker #1: Dan Li from Elsevier
Title: Generative AI in Scientific Knowledge Discovery: Retrieval-Augmented Generation, Agentic Workflows, and Beyond
Abstract: The rapid advancement of generative AI has transformed how researchers interact with scientific knowledge. In this talk, I will present our work at Elsevier on integrating retrieval-augmented generation and multi-agents into research assistants such as Scopus and ScienceDirect to enhance scientific knowledge discovery.
First, I will discuss the development of Scopus AI, a retrieval-augmented generation (RAG) system that leverages large language models for query refinement, document retrieval, and reranking. LLMs are also employed for automated evaluation of generated text and hallucination detection. Next, I will explore how we apply RAG techniques in ScienceDirect to improve content accessibility through automated question generation, on-page reading assistants, and topic page generation. These AI-driven features enable researchers to quickly extract relevant insights from vast bodies of scientific literature.
Throughout the talk, I will highlight key technical components, including LLM-based retrieval, agentic workflows, and knowledge graph integration. By sharing our experiences and challenges, I aim to provide insights into the evolving role of generative AI in the scientific publishing and information access industry.
Bio: Dan Li is a Senior Data Scientist at Elsevier with over 10 years of experience in natural language processing, information retrieval, and LLM-based systems. She leads the development of AI-powered features for products including Scopus AI, LeapSpace, focusing on retrieval-augmented generation, agentic workflows, and trustworthy AI for scientific discovery. Dan holds a PhD in Computer Science from the University of Amsterdam, where her research focused on retrieval models and evaluation for scholarly search.
Speaker #2: Roxana Petcu from University of Amsterdam
Title: T.B.D
Abstract: T.B.D
Bio: T.B.D
SEA talks #297 and #298.18 deelnemers
Verleden evenementen
132
