Yao Lu
Assistant Professor
Works on next-generation AI infrastructure and machine learning systems, and on reinforcement learning for deep research in science.
Yao Lu's lab at NUS works on distributed, cloud & edge systems and reinforcement learning. 15 papers since 2023, 1 at top venues, mostly TMLR. 6 PhD students and 10 master's students. 2 alumni. Applications close .
At a glance
- papers, 2023–2026
- 4/ year
- at T1 venues
- 7%
- PhD students
- 6
- MS students
- 10
- alumni listed
- 2
Papers by yearfrom the lab homepage, 2023–2026
- 8 papers2026 · T1 1 · collecting
- 3 papers2025 · T1 0
- 2 papers2024 · T1 0
- 2 papers2023 · T1 0
T1 venuesOtherVenues: TMLR
Research focuspaper titles that use each topic's words
- Reinforcement learning3 papers
- Distributed, cloud & edge systems1 paper
- next-generation0 papers
- infrastructure0 papers
- learning systems0 papers
Paper list
From the lab homepage.
20268 papers
- The Conference on Empirical Methods in Natural Language Processing (EMNLP). Budapest, HungrayAutomatic Configuration of LLM Post-Training Pipelines (opens in a new tab)
- ACM International Conference on Management of Data (SIGMOD). Bengaluru, India.Efficient LLM Serving for Agentic Workflows: A Data Systems Perspective (opens in a new tab)
- ACM International Conference on AI in Finance (ICAIF) . Milan, ItalyLead-Lag Origination for Prioritizing Financial KOL Narratives
- Proceedings of the Supporting Our AI Overlords workshop at the ACM Conference on AI and Agentic SystemsLumilake: An Agentic Analytics Engine for AI4Science (opens in a new tab)
- TMLR (T1)Planner-Centric Reinforcement Learning for Deep Research with Structure-Aware Reward (opens in a new tab)
- ACM Knowledge Discovery and Data Mining (KDD). Jeju, Korea.Reinforcement Learning with Scenario-Context Rollout in Portfolio Management (opens in a new tab)
- International Conference on Learning Representations (ICLR). Rio de Janeiro, Brazil.Reward Model Routing in Alignment (opens in a new tab)
- International Conference on Learning Representations (ICLR). Rio de Janeiro, Brazil.Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism (opens in a new tab)
20253 papers
- Conference on Empirical Methods in Natural Language Processing (EMNLP) Findings. Suzhou, China.OkraLong: A Flexible Retrieval-Augmented Framework for Long-Text Question Answering (opens in a new tab)
- USENIX Annual Technical Conference (ATC). Boston, MA, USA.Resource Multiplexing in Tuning and Serving Large Language Models (opens in a new tab)
- ACM International Conference on Management of Data (SIGMOD). Berlin, Germany.Scalable Complex Event Processing on Video Streams (opens in a new tab)
20242 papers
- In Proceeding of the Workshop on Machine Learning and Systems.Deferred continuous batching in resource-efficient large language model serving (opens in a new tab)
- In Proceeding of the Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS)UDA: A Benchmark Suite for Retrieval Augmented Generation in Real-world Document Analysis (opens in a new tab)
20232 papers
- Proceedings of the Workshop on Data Management for End-to-End Machine Learning. (DEEM).EVA: An End-to-End Exploratory Video Analytics System (opens in a new tab)
- International Conference on Very Large Data Bases (VLDB). Vancouver, Canada.Serving and Optimizing Machine Learning Workflows on Heterogeneous Infrastructures (opens in a new tab)
Where alumni went2 listed
Apply to NUSFall 2027
School of Computing PhD in Computer Science
- Round
- Main
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USMIT
Song Han
- Efficient training & inference
- USUT Austin
Diana Marculescu
- Accelerators, chips & architecture
- USUIUC
Yongjoo Park
- Efficient training & inference
- CNZhejiang
Kun Yang
- Accelerators, chips & architecture
- USUCSD
Farinaz Koushanfar
- Efficient training & inference
- USUSC
Priyadarshini Panda
- Accelerators, chips & architecture
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
NUScloses