Tutorials

Saturday, March 28, 2026

T1: AI-assisted Scientific Discovery, Experimentation, Content Generation, and Evaluation

  • Organizers: Yufang Hou, Steffen Eger, Anne Lauscher, Wei Zhao and Yong Cao
  • Website: https://ai4science-tutorial.github.io/
  • Description: Expository talks with code running
  • Location: Pavillon DE RABAT (Level 1)
  • Time: Half Day 09:00 - 12:30

Abstract: With the advent of large multimodal foundation models such as ChatGPT, Gemini, Claude, and DeepSeek, scientific research stands at the threshold of an AI-based technological transformation. Recent surveys indicate that the majority of researchers anticipate AI will become mainstream in scientific research within the next two years. This tutorial provides an in-depth overview of recent advances in AI-assisted tools and models that support and enhance the entire scientific research process, building upon findings from our recent survey paper.

We will explore how AI is revolutionizing each stage of the research cycle: (1) Literature Search and Summarization – examining AI-enhanced search systems, paper chat interfaces, graph-based knowledge discovery, and personalized recommender systems; (2) Idea Generation and Experimentation – covering LLM-based hypothesis formulation, multi-agent systems, and automated experimentation tools; (3) Multimodal Content Understanding and Generation – surveying approaches to scientific figure comprehension, automatic diagram generation from text, and poster/slide creation; (4) Text-based Content and Table Generation – reviewing models for abstracts, citations, meta-analysis tables, and long-form content like survey papers; and (5) AI-supported Peer Review – introducing automated review analysis, feedback generation, and meta-review synthesis.

T2: Cognitive Effects and Biases in Large Language Models

  • Organizers: Markus Schedl, Antonela Tommasel, Shahed Masoudian, Ralph Hertwig
  • Website: https://cpjku.github.io/cobis_llm_eacl2026/
  • Description: Talks and hands-on
  • Location: Pavillon DE RABAT (Level 1)
  • Time: Half Day 14:00 - 17:30

Abstract: Cognitive effects such as anchoring, positional effect, or confirmation bias are core aspects of human decision making and reasoning. As LLMs increasingly act as communicative partners, reasoning tools, and evaluators, understanding how these cognitive effects influence their behavior and vice versa has become essential. While recent studies have adapted psychological experiments to detect cognitive biases in LLMs, they often use a particular kind of experimental setup from psychology that carries implications for human performance. In addition, current NLP studies often confuse cognitive effects with biases, diverging from their psychological foundations and overlooking potentially functional aspects of these phenomena.

In this tutorial, jointly organized by NLP researchers and a cognitive psychologist and decision scientist, we aim to build shared conceptual and methodological ground between the two disciplines. We begin by outlining how cognitive effects and biases are defined, validated, and sometimes debated within psychology, highlighting differences and contradictions in experimental designs. We then bridge these insights to NLP through an overview of key studies examining cognitive biases in LLMs, mapping their methodological parallels and divergences. The tutorial also includes a hands-on component where participants explore the challenges of detecting a single cognitive bias (e.g., positional bias) in multilingual LLMs, illustrating the nuances and pitfalls of such evaluations. We conclude by discussing emerging research directions and open questions at the intersection of cognitive science and large language models.

Sunday, March 29, 2026

T3: Encoding and Decoding Language in the Brain with Language Models

Abstract: This EACL 2026 tutorial will cover the foundations of brain–language model alignment and will then explore recent advances on scaling laws of language models for brain alignment, multilingual brain encoding, recent developments in fine-tuning language models with brain data, and advances in brain decoding using language models, including semantic reconstruction of continuous language from brain data. Participants will gain an overview of current naturalistic datasets, computational frameworks, and methods driving the emerging field of NeuroAI. The learning objectives are: (1) Understand the fundamental concepts of brain-AI alignment and encoding models; (2) Learn methodologies for comparing brain activity with model representations; (3) Gain insights into multilingual processing in both human brains and language models; (4) Master techniques for brain-informed model fine-tuning and evaluation; and (5) Discover practical applications and future research directions in Neuro-AI.

T4: Multimodal Large Language Models for Human-AI Interaction: Foundations, Agents, and Inclusive Applications

  • Organizers: Shafiq Joty, Enamul Hoque, Ahmed Masry, Spandana Gella and Samira Ebrahimi Kahou
  • Website: https://mllm4haii.github.io/
  • Description: Expository talks with code running
  • Location: Salle Le Riad (Level 1)
  • Time: Half Day 9:00-12:30

Abstract: Multimodal large language models (MLLMs) are redefining how humans communicate and collaborate with machines. They extend the capabilities of text-based LLMs to perceive, reason, and act across text, images, charts, forms, and graphical user interfaces (GUIs). These models are now capable of answering questions about charts, summarizing infographics, operating software through natural language, and supporting multilingual and accessible visualization.

This tutorial offers a concise, three-hour introduction to the foundations, agentic capabilities, and inclusive applications of MLLMs, with a focus on visually grounded and interactive language tasks. We will cover core architectural designs (encoders, connectors, fusion and decoding mechanisms), multimodal alignment and learning strategies, and reasoning techniques for structured visuals such as charts, forms, and infographics. The tutorial then examines multimodal and conversational agents that perform dialogue-driven reasoning and co-creative analysis in graphical user interfaces. We conclude with discussions on accessibility, multilingual communication, responsible deployment, and future challenges in building human-centered multimodal AI.