Mengdi Wu

4 papers · 2022–2026 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🐣 Hot Topic Early Bird 👥 Mega-Team · Squad

Venues

OSDI (2) ICML (1) NSDI (1)

Top co-authors

Xupeng Miao (3) Zhihao Jia (3) Xinhao Cheng (3) Yingyi Huang (2) Jianan Ji (2) Linfeng Zhang (1) Zepeng Zhao (1) Gabriele Oliaro (1) Remi Delacourt (1) Ziruo Xiao (1)

Research topics

Deep Learning (3) Machine Learning (2) Application Areas (2) Artificial Intelligence (1) Computer Science (1) Applications (1) Core AI (1) Learning Types (1) Optimization & Theory · Deep Learning (1) Optimization & Theory · Machine Learning (1) Systems (1) Techniques (1)

Keywords

tensor program (2) model compression (1) graph representation (1) model serving (1) kernel optimization (1) kernel fusion (1) gaussian mixture model (1) parameter-efficient fine-tuning (1) activation memory (1) weight quantization (1) inference acceleration (1) equivalence verification (1) schedule transformation (1) large language model (1) token scheduling (1) software pipelining (1) persistent kernel (1) multi-gpu inference (1) neural network quantization (1) streaming multiprocessor (1)

Papers

FlexLLM: Token-Level Co-Serving of LLM Inference and Finetuning with SLO Guarantees NSDI 2026 MPK: A Compiler and Runtime for Mega-Kernelizing Tensor Programs OSDI 2026 Mirage: A Multi-Level Superoptimizer for Tensor Programs OSDI 2025 Finding the Task-Optimal Low-Bit Sub-Distribution in Deep Neural Networks ICML 2022