Luo Mai
Reader
Researches machine learning systems across the stack, from models and data to runtimes, systems software and AI hardware.
Luo Mai's lab at Edinburgh works on accelerators, chips & architecture. 13 papers since 2023, 12 at top venues, mostly OSDI and SOSP. 15 PhD students. 3 alumni, most often to NVIDIA and Reka AI.
At a glance
- papers, 2023–2026
- 3/ year
- at T1 venues
- 92%
- PhD students
- 15
- MS students
- 0
- alumni listed
- 3
Papers by yearfrom the lab homepage, 2023–2026
- 6 papers2026 · T1 5 · collecting
- 2 papers2025 · T1 2
- 3 papers2024 · T1 3
- 2 papers2023 · T1 2
T1 venuesOtherVenues: OSDI, SOSP, ICML, HPCA, JMLR
Research focuspaper titles that use each topic's words
- Accelerators, chips & architecture2 papers
- researches machine0 papers
- learning systems0 papers
- hardware0 papers
- software0 papers
Paper list
From the lab homepage.
20266 papers
- OSDI (T1)BatchGen: An Architecture for Scalable and Efficient Batch Inference
- HPCA (T1)BitDecoding: Unlocking Tensor Cores for Long-Context LLMs with Low-Bit KV Cache
- MLSys (T1)ContextPilot: Fast Long-Context Inference via Context Reuse
- SOSP (T1)MeshRT: Compile-Time Governed Wafer-Scale Runtime for Low-Latency High-Throughput Inference
- ACL DemoRyze: Evidence-Enriched Data Synthesis from Biomedical Papers
- SOSP (T1)Wavel: A Fast and Efficient Compilation System for Wafer-Scale Accelerators
20252 papers
- NeurIPS (T1)MoE-CAP: Benchmarking Cost, Accuracy and Performance of Sparse Mixture-of-Experts Systems
- OSDI (T1)WaferLLM: Large Language Model Inference at Wafer Scale
20243 papers
- ICML (T1)Learning high-frequency functions made easy with sinusoidal positional encoding
- OSDI (T1)ServerlessLLM: Low-Latency Serverless Inference for Large Language Models
- SOSP (T1)Tenplex: Dynamic Parallelism for Deep Learning using Parallelizable Tensor Collections
20232 papers
- ICML (T1)GEAR: A GPU-Centric Experience Replay System for Large Reinforcement Learning Models
- JMLR (T1)TorchOpt: An Efficient Library for Differentiable Optimization
Where alumni went3 listed, 100% to industry
Industry 3Academia 0
Most often to
- NVIDIA
- Reka AI
- Tencent
Apply to EdinburghFall 2027
Not announced yet
Informatics PhD
- Round
- Rolling (funding consideration)
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- none
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USUCLA
Hanrui Wang
- Accelerators, chips & architecture
- USUMD
Cunxi Yu
- Accelerators, chips & architecture
- USStanford
Priyanka Raina
- Accelerators, chips & architecture
- CNTsinghua
Xing Lin
- Accelerators, chips & architecture
- DETUM
Hussam Amrouch
- Accelerators, chips & architecture
- CNTsinghua
Peng Qu
- 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.