Possible advisorsa guess from early papers, not confirmed

• Anh Tuan Luu

  Possible advisor · last author on 7 of their early first-author papers, 2025–2026

  Suggested from co-authorship

Is this you? Claim this profile to confirm or dismiss it.

• From Signals to Transfer: A Factorised Study of Probe-Based Uncertainty Estimation in Large Language Models

  Ponhvoan Srey, X . Wu, Cong-Duy T Nguyen, Quang Minh Nguyen, Duc Anh Vu, Anh Tuan Luu

  arXiv · 2026

  The results show that raw hidden states and attention features are difficult to outperform in-domain, however, under distribution shift, structured and compressed features are more robust, suggesting that in-domain performance alone is insufficient to measure progress.

  2
• TIGER: Text-Conditioned Visual Gated Routing with Acceptance Alignment for Multimodal Speculative Decoding

  Quynh Vo, Cong-Duy T Nguyen, Ponhvoan Srey, Luu Anh Tuan, Thong Nguyen

  arXiv · 2026

  TIGER, a Text-conditioned vIsual GatEd Routing framework for multimodal speculative decoding, is proposed, which yields consistent gains in accepted prefix length and speculative speedup under exact verifier-side speculative decoding, while achieving favorable quality-latency trade-offs with comparable downstream accuracy in visual-routing analyses.

  2
• Uncover and unlearn nuisances: agnostic fully test-time adaptation

  Ponhvoan Srey, Yaxin Shi, Hangwei Qian, Jing Li, Ivor Wai-Hung Tsang

  Machine Learning · 2025

  This work exploits a dual perspective on FTTA, and proposes Agnostic FTTA (AFTTA) as a novel formulation that enables the usage of off-the-shelf domain transformations during test-time to enable direct generalization to unforeseeable target data.

  1
• Unsupervised Hallucination Detection by Inspecting Reasoning Processes

  Ponhvoan Srey, Xiaobao Wu, Anh Tuan Luu

  Conference on Empirical Methods in Natural Language Processing (EMNLP) · 2025

  1
• Learning Uncertainty from Sequential Internal Dispersion in Large Language Models

  Association for Computational Linguistics 2026, Anh Tuan Luu, Ponhvoan Srey, Cong-Duy T Nguyen, Xiaobao Wu

  Underline Science Inc. · 2026

  –
• Towards Reliable Truth-Aligned Uncertainty Estimation in Large Language Models

  Ponhvoan Srey, Quang Minh Nguyen, Xiaobao Wu, Anh Tuan Luu

  arXiv · 2026

  This work proposes Truth AnChoring (TAC), a post-hoc calibration method to remedy UE metrics, by mapping the raw scores to truth-aligned scores, and shows that UE metrics become non-discriminative precisely in low-information regimes.

  –
• Reinforcing Step-level Reasoning for Effective Self-Correction in LLMs

  Vu Duc Anh, Nhat M. Hoang, Do Xuan Long, Cong-Duy T Nguyen, Ponhvoan Srey, Luu Anh Tuan

  arXiv · 2026

  –
• Open-World Learning Under Dataset Shift

  Ponhvoan Srey, Yuhui Zhang, Takafumi Kanamori

  IEEE Conference on Artificial Intelligence (CAI) · 2024

  –