Yifan Wang

Yifan Wang

Assistant Professor in University of Hawaii at Manoa

Posts by Tags

AI

Next Generation AI-Enhanced Multi-Modal Database Optimizer

less than 1 minute read

Published:

We develop the next generation multi-modal query optimizer with the latest AI techniques. In our recent paper LaPuda: LLM-Enabled Policy-Based Query Optimizer for Multi-Modal Data, a large language model (LLM) based query optimizer is designed for efficient multi-modal, semantic query processing.

LaPuda architecture

Combining powerful LLM and traditional database optimization techniques, the new optimizer achieves significant performance to optimize the multi-modal query plans (with both structured and semantic information, covering multiple data modalities), resulting in 3x query execution speedup for complex multi-modal queries. The design scheme also proves the effectiveness of using LLM in modern query optimizers, especially for complex semantic queries. This project is an important step towards practical design of multi-modal data systems in the LLM era.

Algorithm

Leetcode # 201-250

2 minute read

Published:

My solutions to some Leetcode problems ranging in #201-250.

C++

C++ Notes (1)

2 minute read

Published:

Notes on C++ (1).

Compiler

Some Notes on ASM

4 minute read

Published:

ASM is an important tool for manipulating java bytecode. Here is a summary of typical problems when I was implementing a compiler using Java and ASM.

Database

Next Generation AI-Enhanced Multi-Modal Database Optimizer

less than 1 minute read

Published:

We develop the next generation multi-modal query optimizer with the latest AI techniques. In our recent paper LaPuda: LLM-Enabled Policy-Based Query Optimizer for Multi-Modal Data, a large language model (LLM) based query optimizer is designed for efficient multi-modal, semantic query processing.

LaPuda architecture

Combining powerful LLM and traditional database optimization techniques, the new optimizer achieves significant performance to optimize the multi-modal query plans (with both structured and semantic information, covering multiple data modalities), resulting in 3x query execution speedup for complex multi-modal queries. The design scheme also proves the effectiveness of using LLM in modern query optimizers, especially for complex semantic queries. This project is an important step towards practical design of multi-modal data systems in the LLM era.

Distributed Systems

Elixir

Java

Some Notes on ASM

4 minute read

Published:

ASM is an important tool for manipulating java bytecode. Here is a summary of typical problems when I was implementing a compiler using Java and ASM.

Leetcode

Leetcode # 201-250

2 minute read

Published:

My solutions to some Leetcode problems ranging in #201-250.

Query Optimization

Next Generation AI-Enhanced Multi-Modal Database Optimizer

less than 1 minute read

Published:

We develop the next generation multi-modal query optimizer with the latest AI techniques. In our recent paper LaPuda: LLM-Enabled Policy-Based Query Optimizer for Multi-Modal Data, a large language model (LLM) based query optimizer is designed for efficient multi-modal, semantic query processing.

LaPuda architecture

Combining powerful LLM and traditional database optimization techniques, the new optimizer achieves significant performance to optimize the multi-modal query plans (with both structured and semantic information, covering multiple data modalities), resulting in 3x query execution speedup for complex multi-modal queries. The design scheme also proves the effectiveness of using LLM in modern query optimizers, especially for complex semantic queries. This project is an important step towards practical design of multi-modal data systems in the LLM era.

WinAPI

C++ Notes (1)

2 minute read

Published:

Notes on C++ (1).