Schedule
LLAIS Workshop at ECAI 2025
09:00 – 9:10
Welcome and overview by organizers
09:10 – 9:35
Meysam Asgarichenaghlou
Invited Speaker
09:35 – 10:10
Duel-RAG: Retrieval-Augmented Generation for Yu-Gi-Oh!
Sirui Zhu and Charlie Cheng
Improving LLM Agents with Reinforcement Learning on Cryptographic CTF Challenges
Lajos Muzsai, David Imolai and András Lukács
10:10 – 10:30
Balancing Adaptivity and Alignment: A Multi-Agent System for Personalized LLM Judge design
Hongliu Cao, Ilias Driouich, Robin Singh and Eoin Thomas
VIRAL: Vision-grounded Integration for Reward design And Learning
Valentin Cuzin-Rambaud, Emilien Komlenovic, Alexandre Faure and Bruno Yun
LLM-based intelligent agents for sustainable last-mile logistics: Insights from parcel locker use and expert evaluation
Anna Drab-Kurowska, Agnieszka Budziewicz-Guźlecka, Joanna Drobiazgiewicz, Piotr Gutowski, Michał Kuściński and Krzysztof Lorenz
10:30 – 10:45
10:45 – 11:15
Purvanshi Mehta
Invited Speaker
11:15 – 12:22
ID-RAG: Identity Retrieval-Augmented Generation for Long-Horizon Persona Coherence in Generative Agents
Daniel Platnick, Mohamed Elamine Bengueddache, Marjan Alirezaie, Dava J. Newman, Alex Sandy Pentland and Hossein Rahnama
Analyzing the Correlation Between Hallucinations and Knowledge Conflicts in Large Language Models
Lucrezia Laraspata, Giovanna Castellano and Gennaro Vessio
A Graph-Enhanced LLM-Based Question Answering System for the AI Act
Nicola Aggio, Alberto De Lazzari and Teresa Scantamburlo
D-LEXI-LLM: A Depression-aware and Lexically-Guided Language Model for Analyzing Social Media Posts
Farzaneh Lashgari, Sebastião Pais, Mehran Pourvahab, António Sousa and Anilson Monteiro
12:22 – 12:30
Evaluating Representational Fairness in Multilingual LLMs via Cloze-Based Trait Attribution
Saba Ghanbari Haez, Simone Magnolini, Monica Consolandi and Mauro Dragoni
12:30 – 12:35
Summary, acknowledgments, and announcements