Edoardo Ponti
Top university
Just Landed
Posted Yesterday
Postdoc and Fully Funded PhD Positions in Adaptive Tokenization and Memory in Foundation Models at Imperial College London Imperial College London in United Kingdom
Degree Level
PhD, Postdoc
Field of study
Computer Science
Funding
The positions are fully funded. The post mentions project-owned B200 GPUs and cloud credits, departmental H200s, funding for two conferences per year, and a 2-year Research Associate postdoc. The PhD students are described as fully funded and approximately 3.5 years in duration.
Deadline
Sep 30, 2026
Country
United Kingdom
University
Imperial College London
How do I apply for this?
Sign in for free to reveal details, requirements, and source links.
Apply for this position
Keywords
Suggested positions
About this position
Imperial College London is recruiting for research positions in adaptive tokenization and memory in foundation models as part of Edoardo Ponti’s ERC Starting Grant project, AToM (Adaptive Tokenization and Memory in Foundation Models).
The call includes 1 postdoc / Research Associate and 2 fully funded PhD students. The research agenda focuses on latent tokenization, permanent adaptive-size memory, hierarchical sparse attention, retrofitting state-of-the-art models, and applications in inference-time hyper-scaling, reasoning, agentic workflows, and long-horizon world modelling.
The lab highlights prior work such as Dynamic Token Pooling, Bolmo, and Dynamic Memory Compression, with open-weight model releases in collaboration with NVIDIA and Ai2. Successful candidates will work with project-owned B200 GPUs, departmental H200s, cloud credits, and receive funding for two conferences per year.
The postdoc is a 2-year Research Associate role with flexibility to set an independent research agenda and co-supervise PhD/MSc students. The PhD positions are approximately 3.5 years and are applied for through Imperial Computing admissions on a rolling basis.
Deadline: 30 September 2026 for the postdoc. The PhD applications are rolling.
How to apply: Use the expression-of-interest link for the postdoc and the Imperial Computing admissions route for the PhD positions. Review the details link for full instructions.
Funding details
The positions are fully funded. The post mentions project-owned B200 GPUs and cloud credits, departmental H200s, funding for two conferences per year, and a 2-year Research Associate postdoc. The PhD students are described as fully funded and approximately 3.5 years in duration.
What's required
Applicants should be interested in research on foundation models, latent tokenization, adaptive memory, hierarchical sparse attention, multimodal learning, model retrofitting, and inference-time scaling. The postdoc is a 2-year Research Associate role with freedom to set an independent research agenda and co-supervise PhD/MSc students. PhD applicants should apply through Imperial Computing admissions on a rolling basis; no specific degree, GPA, or language-test requirements are stated in the post.
How to apply
For the postdoc, submit an expression of interest using the provided link. For the PhD positions, apply through Imperial Computing admissions on a rolling basis. Check the details link for full information before applying.
More information can be found here
Ask ApplyKite AI
Professors
How do I apply for this?
Sign in for free to reveal details, requirements, and source links.