全奖博士丨英国巴斯大学招收大型语言模型方向全奖博士

项目简介

This project is one of a number that are in competition for funding from the University of Bath URSA competition: bath.ac.uk/campaigns/university-research-studentship-award-ursa/, for entry in September 2025.

Supervisory Team:

Dr. Da Chen & Dr Yongliang Yang

Overview of the Research:

Recent years have witnessed remarkable developments in Large Language Models and further expansion to Large Multimodal Models (with text, audio, image, video, etc.). However, challenges arise when models need to “unlearn” certain aspects of their training data when it comes to applications in sensitive domains like healthcare, law enforcement, and finance. Particularly, with laws such as GDPR, it is important to ensure that models can unlearn data for individuals. In these high-stakes environments, it is critical that models can effectively “unlearn” specific data points that may be private, incorrect, outdated, biased, or legally restricted.

Unlearning refers to the ability of models to forget or remove specific knowledge that may have been unintentionally biased, incorrect, or no longer relevant while retaining other knowledge. This is challenging as these types of samples are normally insufficient in terms of the numbers, annotations, and complexities.

In this project, we will explore:

1. Unlearning Mechanisms with limited data: Developing algorithms and methods that allow large multimodal models to unlearn specific pieces of data efficiently and selectively with limited target samples provided.

2. Model generalization: Investigating how unlearning impacts the integrity, consistency, and performance of large multimodal models across tasks.

3. Applications in Privacy and Bias: Apply the proposed method for applications such as data privacy, bias mitigation, and ethical AI, especially when sensitive data must be removed from the pre-trained large models.

Support and Resources:

You will be part of the Visual Computing Group (VCG) and Artificial Intelligence & Machine Learning group(AI&ML) within the Department of Computer Science which is highly regarded internationally and very successful in research publications. Our group is affiliated to a £10 Million EPSRC funded CAMERA research centre with state-of-the-art studio facility and a great and active research cohort with more than 20 Ph.D. students

We work closely with collaborators from related industries and develop high-quality, impactful and practical research.

Project keywords: Unlearning, Large Language Models, Large Multimodal Models, Multimodal Learning, Few-shot Learning, Long-tailed Data Learning, medical images, privacy, ethical AI.

申请要求

Applicants should hold, or expect to receive, a First Class or good Upper Second Class UK Honours degree (or the equivalent) in a relevant subject. A master’s level qualification would also be advantageous.

Non-UK applicants must meet the programme’s English language requirement: https://www.bath.ac.uk/corporate-information/postgraduate-research-degrees-english-language-requirements-for-international-students/ prior to the closing date of this advert.

The project is open to home students and to exceptional overseas applicants. An explanation of what is meant by Home student status is given on our website: https://www.bath.ac.uk/guides/understanding-your-tuition-fee-status/

申请方式

Informal enquiries are encouraged and should be directed to supervisor Dr. Da Chen: https://researchportal.bath.ac.uk/en/persons/da-chen, dc598@bath.ac.uk

Formal applications should be submitted via the University of Bath’s online application form: https://samis.bath.ac.uk/urd/sits.urd/run/siw_ipp_lgn.login?process=siw_ipp_app&code1=RDUCM-FP01&code2=0019 for a PhD in Computer Science prior to the closing date of this advert.

IMPORTANT:

When completing the application form:

1. In the Funding your studies section, select ‘University of Bath URSA’ as the studentship for which you are applying.

2. In the Your PhD project section, quote the project title of this project and the name of the lead supervisor in the appropriate boxes. 

Failure to complete these two steps will cause delays in processing your application and may cause you to miss the deadline.

More information about applying for a PhD at Bath may be found on our website: https://www.bath.ac.uk/guides/how-to-apply-for-doctoral-study/.

PLEASE BE AWARE: Applications for this project may close earlier than the advertised deadline if a suitable candidate is found. We therefore recommend that you contact the lead supervisor prior to applying and submit your formal application as early as possible.

Equality, Diversity and Inclusion:

We value a diverse research environment and aim to be an inclusive university, where difference is celebrated and respected. We welcome and encourage applications from under-represented groups.

If you have circumstances that you feel we should be aware of that have affected your educational attainment, then please feel free to tell us about it in your application form. The best way to do this is a short paragraph at the end of your personal statement.

奖学金详情

Candidates may be considered for a University of Bath studentship tenable for 3.5 years. Funding covers tuition fees, a stipend (£19,237 p/a in 2024/5) and access to a training support budget.  

近期推荐