Sign up with

Already have an account? Log in here

Need some help?
Talk to us at +91 7670800001

ICL_Vision-Language Model Developmen t Intern (W100) at ITRI

, , -

Full Time

Start Date

Immediate

Expiry Date

05 May, 26

Salary

0.0

Posted On

04 Feb, 26

Experience

0 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

Data Collection, Data Integration, Data Cleaning, Data Processing, Data Annotation, Semantic Annotation, Data Verification, Model Training, Python, Deep Learning, TensorFlow, PyTorch

Industry

Research Services

Description

This internship is in ITRI's Information and Communications Research Laboratories(ICL). Intern will work on the following topics / tasks: Data Collection and Integration: Assist in collecting and integrating visual-language model (VLM) training data, including images, text, and corresponding descriptions, ensuring data diversity and quality. Data Cleaning and Processing: Clean and format collected data to meet training requirements and ensure consistency. Data Annotation: Perform semantic annotation of images and text according to established guidelines, supporting the accuracy of cross-modal learning. Data Verification: Check the accuracy of annotated data, assist in identifying and correcting potential errors, and improve data quality. Model Training Support: Assist in running and testing pre-trained VLM models, provide feedback based on the data characteristics, and contribute to model optimization. Qualifications: Current undergraduate or graduate student, preferably in Computer Science, Artificial Intelligence, or related fields. Proficiency in Python programming. Basic understanding of the data annotation process. Prior experience in image or text annotation is a plus, with the ability to follow annotation guidelines and ensure accuracy. Familiarity with deep learning frameworks (e.g., TensorFlow, PyTorch) is a plus.

Responsibilities

The intern will assist in collecting and integrating visual-language model training data, ensuring data diversity and quality. They will also perform data cleaning, annotation, verification, and support model training.