Embedded Files
I am actively looking for a Research Internship for Spring/Summer 2027.
If you have an open position that aligns with my experience, please reach out!
π About me
π About me
Iβm a second-year PhD student at Virginia Tech, advised by Dr. Tu Vu. I interned at Adobe Research in Summer 2025, where I worked closely with Dr. Viet Lai.
Previously, I got my M.Phil in Computer Science and B.Sc in Data Science and Pure Math (Advanced Track) from HKUST in Sep 2024 and July 2022. During my MPhil program, I was fortunate to be supervised by Prof. Yangqiu Song.
If youβre interested, here is my Curriculum VitaΓ©. Welcome to my homepage!Β
π¬ Research Interest
I aspire to develop AI systems that evolve through efficient self-learning and practical experience, mirroring the trajectory of human intelligence. In the era of large-scale models, my research focuses on:
π Long-context searching: Advancing multi-hop reasoning in LLMs for search applications (Ο^2).
π₯ Post-training: Designing data synthesis pipelines to fine-tune LLMs for various skills and behaviors.
π Agentic memory and context engineering: Designing, evaluating, and improving agentic systems for realistic tasks that require long-horizon interactions and memory.
ποΈ News
Just migrate my homepage from the Github Academic Pages template to Google Sites
[Archived]
π Preprints
(No recent preprint)
π Selected Publications
For an up-to-date list of my research papers, please see my Google Scholar profile. * denotes equal contribution.
[COLM 2026] Ο^2: Structure-Originated Reasoning Data Improves Long-Context Reasoning Ability of Large Language Models
[COLM 2026] Ο^2: Structure-Originated Reasoning Data Improves Long-Context Reasoning Ability of Large Language Models
Quyet V. Do, Thinh Pham, Nguyen Nguyen, Sha Li, Pratibha Zunjare, Tu Vu
// A fully open-source realistic QA generation pipeline that SFT-ing on gpt-oss-20b gains an average of +4.3% absolute accuracy on 4 benchmarks: SealQA, LongBench-v2, Oolong-Synth, OfficeQA[WACV 2026] Vision Language Models Learn to Assess Images with Specialists
[WACV 2026] Vision Language Models Learn to Assess Images with Specialists
Quyet V. Do, Seunghyun Yoon, Ruiyi Zhang, Thiloshon Nagarajah, Trung Bui, Viet Dac Lai
// Improve GPT-4o/Gemini-2.5-Pro by up to 9% and train InternVL3-8B to be on par with them on image assessment.[pdf] [code]. WVAQ Workshop @ WACV2026.
[AACL 2023] A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity
[AACL 2023] A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity
Yejin Bang, Samuel Cahyawijaya, Nayeon Lee, Wenliang Dai, Dan Su, Bryan Wilie, Holy Lovenia, Ziwei Ji, Tiezheng Yu, Willy Chung, Quyet V. Do, Yan Xu, Pascale Fung
// First empirical study on ChatGPTπ Internship
Adobe Research, San Jose, CA, USA.
Research Scientist Intern @ NLG Team, May. 2025 β Aug. 2025
Mentor: Dr. Viet Lai, Dr. David Yoon, and Dr. Trung Bui
Summary: Work on automatic evaluation for image generation and editing. Improve the performance of GPT-4o and Gemini-2.5-Pro in the task by 9% accuracy and Pearson correlation.
π§ Contacts
Feel free to reach me at {quyet}{my last name} {at} vt [dot] edu.
If you want to have a chat with me, please book a Google Meet via this link.
This website is migrated from https://dovanquyet.github.io/
Page updated
Google Sites
Report abuse