Hello there 🍀

Hi, and welcome to my page! I’m Khanh-Tung Tran (I usually go by Tung), a third-year Ph.D. student in Computer Science at University College Cork, Ireland, funded by Research Ireland’s CRT-AI programme. I am supervised by Assoc. Prof. Harry Nguyen and Prof. Barry O’Sullivan. My research focuses on cross-lingual alignment and multilingual capability of large language models (LLMs), particularly their applications in low-resource scenarios to promote fair and inclusive LLM development. I am also interested in agentic AI, especially (collective) multi-agent systems for developing real-world, collaborative AI applications.

News

• 02-2026: 2 of the papers with our collaborators have been accepted to LREC 2026. Congratulations to all involved!
• 02-2026: Our paper, ToolBrain, has been accepted to the Demonstration Program, AAMAS 2026.
• 01-2026: Our paper, LaCoMSA: Language-Consistency Multilingual Self-Alignment with Latent Representation Rewarding, has been accepted to EACL 2026 (main conference).
• 11-2025: Our paper, IRLBench: A Multi-modal, Culturally Grounded, Parallel Irish-English Benchmark for Open-Ended LLM Reasoning Evaluation, has been accepted to KDD 2026 - Datasets & Benchmarks Track.
• 11-2025: Our paper, Reasoning Transfer for an Extremely Low-resource and Endangered Language: Bridging Languages through Sample-Efficient Language Understanding, has been accepted to AAAI-26.
• 11-2025: Presented our paper Disentangling Language Understanding and Reasoning Structures in Cross-lingual Chain-of-Thought Prompting at EMNLP2025.