For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
万少华

万少华

Intern - Data Preprocessing for Multimodal LLM Pretraining

China flagShenZhen, China

Key Skills

Software

No software listed

Top Subject Matter

Multimodal AI (Image + Text)
Industrial Quality Control (Computer Vision)

Top Data Types

ImageImage

Top Task Types

Data CollectionData Collection
Object DetectionObject Detection

Freelancer Overview

Intern - Data Preprocessing for Multimodal LLM Pretraining. Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. AI-training focus includes data types such as Image and labeling workflows including Data Collection and Object Detection.

Labeling Experience

AI Algorithm Engineer - Data Annotation Coordination for Computer Vision

ImageImageObject DetectionObject Detection

I led the development of industrial product defect detection algorithms using labeled image data. I organized and supervised the data annotation workflow, focusing on accurate object labeling and fine-grained defect identification. My efforts contributed to significant improvements in model accuracy for manufacturing quality control. • Managed data annotation tasks for industrial product images. • Optimized annotation guidelines for micro-defect detection. • Coordinated between annotation teams and engineers for iterative feedback. • Enhanced object detection accuracy through meticulous label quality assurance.

2023 - Present

Intern - Data Preprocessing for Multimodal LLM Pretraining

ImageImageData CollectionData Collection

I participated in the pretraining data cleaning process for a multimodal large language model using image-text pair data. I developed and maintained automated data filtering scripts to screen over 1 million image-text pairs for quality assurance. The work improved the dataset acceptance rate, ensuring a higher standard for subsequent model training. • Conducted quality checks and filtering on over 1 million image-text pairs. • Improved qualified data rate from 90% to 98% through automated tooling. • Supported model training by providing high-quality annotated data. • Collaborated with the AI development team to establish data selection criteria.

2022 - 2023

Education

深圳职业技术大学

本科, 多媒体设计

本科
2017 - 2021

Work History

I

Iflytek Artificial Intelligence Technology

Algorithm Intern

Hangzhou
2022 - 2023