PROGRAM

27 January 2026, Singapore Expo (room number TBA)

Session 1 (Chair: Michal Ptaszynski)

08:30 - 08:45 Bin Han - Personality Expression Across Contexts: Linguistic and Behavioral Variation in LLM Agents

08:45 - 09:00 Naoki Takada - Rethinking Dialogue Disentanglement for LLMs via Dialogue-Level Assignment and Subsequent Context

09:00 - 09:15 Soyeong Jeong - Adaptive Multi-Agent Response Refinement in Conversational Systems

Session 2 (Chair: Pawel Dybala)

09:15 - 09:30 Rene Melendez - PhishLLM: Generating Japanese Phishing Emails to Test Cross-Lingual Generalization in English-Trained Models

09:30 - 09:45 Yuta Nakajima - End-to-End Assessment of Product Review Helpfulness Using Subjective and Objective Information

Break / Poster session (9:45-11:15)

Hsien-Te Kao - Evaluating LLM Alignment under Big Five Personality Prompting

Ratna Kandala - Evaluating Adaptive AI Support: Human Perceptions of LLM-Generated Relationship Advice

Ziqi Zhu - Conversational Intent-Driven GraphRAG: Enhancing Multi-Turn Dialogue Systems through Adaptive Dual-Retrieval of Flow Patterns and Context Semantics

Murilo da Luz - Partial Reasoning in Language Models Search and Refinement Guided by Uncertainty

James Hale - Can LLMs Mediate Synchronous Dispute Dialogues?

Rafal Rzepka - Evaluating Lightweight Embedding Guardrails for Cost-Effective Misalignment Mitigation in Export Control Dialog Systems

Shinji Muraji - Evaluation of Emotion-Conditioned Response Generation for Role-Playing Agents Using Large Language Models: A Case Study with Facial Expression Labels from Visual Novel Data

Session 3 (Chair: Siaw-Fong Chung)
11:15 - 11:30 Hassan Hamad - ToolCritic: Detecting and Correcting Tool-Use Errors in Dialogue Systems (withdrawn)

11:15 - 11:30 Nirmalie Wiratunga - iCARE: Ontology-Guided Intent Routing for Multi-Agent LLM-Based Dialogue Systems

11:30 - 11:45 Hiroshi Shigenobu - A Novel Neuro-symbolic Approach to Irony Detection Based on Structural Components of Ironic Statements

Invited Talk (Chair: Rafal Rzepka)
11:45 - 12:30 Rui Mao - Metaphorical Cognition and Its Computational Practices

Abstract: Understanding human cognitive structures has long been a challenging endeavor. The methodological tools available to us remain relatively limited. Metaphorical cognition, however, offers a promising avenue forward. Far from being a decorative feature of language, metaphor constitutes a core cognitive mechanism through which individuals organise concepts, interpret abstract domains, and make decisions. In this talk, I present a cognitively grounded account of metaphor that unites the principles of Conceptual Metaphor Theory with modern computational techniques. Building on recent developments in our MetaPro framework, I will demonstrate how large-scale metaphor processing can uncover latent cognitive patterns embedded in discourse across domains such as AI, finance, public health, and mental well-being. I will also highlight emerging evidence that metaphorical framing produces quantifiable cognitive signatures, reflected in both neural activity and behavioural measures. By integrating humanistic insight with data-driven analysis, this talk highlights the scientific significance and practical potential of computational metaphor research for advancing our understanding of how people think, feel, and act through language.

Invited Speaker BIO

Dr. Rui Mao is a Research Scientist and Lead Investigator in Artificial Intelligence at Nanyang Technological University. His work spans cognitive computing, affective computing, and quantitative finance, with a particular emphasis on understanding how human cognitive mechanisms can be operationalised in computational systems. He is the developer of MetaPro, the first fully end-to-end framework for large-scale computational metaphor processing. MetaPro has become an influential research instrument, enabling systematic cognitive analysis across diverse domains, including artificial intelligence, public health, mental well-being, finance, linguistics, and marketing, and supporting both theoretical inquiry and real-world applications.

Links to past LaCATODA workshops:

Page updated

Google Sites

Report abuse