Program

Welcome Remarks

Keynote Talk 1: Pluralism and the Paradox of AI Alignment: Rethinking Human Intelligence in the Age of AI (Rostam J. Neuwirth, University of Macau)

Oral Session 1

1. 09:40–09:52 EduValueBench: Benchmarking Situated Value Alignment of LLMs in Chinese Educational Decision-Making Junru Wu
2. 09:52–10:06 Denser, Not Wider: Training-Data Structure Selects the Kind of Pluralism a Fine-Tuned Model Retains Aayush Singh
3. 10:06–10:18 Effects of Post-Training on Cultural Anchoring in LLMs Alex Chandy
4. 10:18–10:30 Guided Relational Analysis Improves Social Intelligence in LLMs Anvesh Rao Vijjini

Coffee Break

Keynote Talk 2: Whose Values, Whose Safety? Building LLM Safety for a Pluralistic World (Roy Ka-Wei Lee, University of British Columbia)

Oral Session 2

1. 11:35–11:47 StealthRL: Reinforcement Learning Paraphrase Attacks for Multi-Detector Evasion of AI-Text Detectors Suraj Ranganath
2. 11:47–11:59 Performative Alignment: Preference Learning under Deployment-Dependent Utility Tarun Chinta
3. 11:59–12:11 Country Name or Cultural Evidence? A Causal Audit of Population-Grounded Steering in Large Language Models Ziyan Yang
4. 12:11–12:23 NepPlural: A Human-Anchored Benchmark for Persona Annotation in Nepali Migration Discourse Jinu Nyachhyon

Best Paper Award — Country Name or Cultural Evidence? A Causal Audit of Population-Grounded Steering in Large Language Models (Ziyan Yang)

Lunch

Keynote Talk 3: Grounding AI in Society (Nancy F. Chen, Center for Frontier AI Research, A*STAR)

Oral Session 3

1. 14:35–14:47 Findings of the PlurVA-LLM 2026 Shared Task on Pluralistic Value Alignment Across China, Indonesia, and Sri Lanka Supryadi
2. 14:47–14:59 PlurVA-LLM-2026 Shared Task Track-1 and 2: Preference Learning for Pluralistic Value Alignment Joydeb Mondal
3. 14:59–15:11 PlurVA-LLM-2026 Shared Task Track-1: Pluralistic Value Alignment in LLMs via Multilingual Fine-Tuning and Threshold Calibration Nevidu Jayatilleke
4. 15:11–15:23 MasalBench: A Benchmark for Contextual and Cross-Cultural Understanding of Persian Proverbs in LLMs Ghazal Kalhor

Best System Award — PlurVA-LLM-2026 Shared Task Track-1 and 2: Preference Learning for Pluralistic Value Alignment (Joydeb Mondal)

Coffee Break

Keynote Talk 4: Pluralistic Alignment Always, Sometimes, or Never? Taking Value Pluralism Seriously (Pak-Hang Wong, Hong Kong Baptist University)

Poster Session

1. IndRegBias: A Dataset for Studying Indian Regional Biases in English and Code-Mixed Social Media Comments Akash Anil
2. Concepts at Work: Testing Rival Conceptions of Liberty in Large Language Models Mark J. Hill
3. The Language You Ask In: Language-Conditioned Ideological Divergence in LLM Analysis of Contested Political Documents Oleg Smirnov
4. C-VoicesCompass: Evaluating Value Alignment of LLMs with Chinese Social Values Yuemei Xu
5. CompOrca: Corpus-Scale Compliance Labelling of Instruction-Tuning Data Philipp E. Glass
6. PlurVA-LLM-2026 Shared Task Track-2: PluriGuide—Principle Distillation and Identifier-Bound Demonstrations for Pluralistic Value Alignment Junyu Lu
7. PlurVA-LLM-2026 Shared Task Track-1: Curriculum Fine-Tuning with Soft Labels and Positional Consistency for Cross-Cultural Value Alignment Nevidu Jayatilleke
8. Pancasila-Dilemmas: Evaluating Large Language Models on Indonesian Human Value Dilemmas Grounded in Pancasila Supryadi
9. Aligning Large Language Models with Pluralistic Values: Case Study in Southeast Asian (SEA) Countries Supryadi
10. ABench: Adversarily Benchmarking Large Language Models across Taxonomized Values Liping Yang
11. LKValues: Aligning Large Language Models with Sri Lankan Societal Values Nethmi Muthugala
12. Beyond Value Benchmarks: Measuring Value-Structure Alignment in Large Language Models via Symmetric Q-Sorts Jingting Zheng
13. VALCON: Benchmarking and Enhancing Value Consistency in Large Language Models via Dual-Role Variation Junru Wu
14. The Alignment Problem Is an Identity Problem: Lessons from Four Billion Years of Major Evolutionary Transitions Nyx Redondo
15. Auditing LLM-Governed Social Robots with Culture-Specific Moral Gradients Carmen Ng
16. One Prompt Does Not Fit All: Self-Meta-Evolve for Personalized Information Extraction Hongliang Li

Closing Session