Research Proposals for Faculty

The JHU + Amazon Initiative for Interactive Artificial Intelligence (AI2AI) solicits research proposals from faculty for advancing the state of the art in all aspects of interactive AI.

Amazon builds and deploys AI across three technology layers: the bottom layer consists of Amazon’s own high performance and cost-effective custom chips, as well as a variety of other computing options including from third-parties. The middle layer focuses on the customer’s choice by providing the broadest selection of Foundation Models—both Amazon-built as well as those from other leading providers. At the top layer Amazon offers generative AI applications and services to improve every customer experience.

There are three things that distinguish Amazon’s approach to the development and deployment of AI:
1) Maintaining a strategic focus on improving the customer and employee experience through practical, real-world applications of AI;
2) marshaling world-class data, compute, and talent resources to drive AI innovation; and
3) committing to the development of responsible, reliable, and trustworthy AI.

Topics of interest include, but are not limited to, those below. Please feel free to bring your own unique viewpoint and expertise to these topics:

Topics

  • Foundation Model Improvements:
  • Novel model architectures
  • Novel training algorithms and methodologies
  • Multi-modal (e.g., text, image, video, audio) understanding and generation
  • Multi-lingual understanding and generation
  • Algorithms and workflows for acquiring and curating high-quality and diverse datasets for training
  • Foundation Model Evaluation:
  • Creation of new benchmarks for assessing foundation model capabilities
  • Methodologies for robust evaluation of generative AI systems, including agents
  • Efficient Generative AI:
  • Efficient training and inference for cloud and on-device applications
  • Compute and memory efficient handling of multi-modal long/infinite context
  • Improving efficiency of diffusion models, including discrete diffusion
  • Reasoning:
  • Commonsense and domain-specific (e.g., math, coding) reasoning
  • Temporal and spatial reasoning
  • Reasoning for planning
  • Test-time scaling
  • Creative problem solving (learning out-of-the-box thinking techniques)
  • Knowledge Grounding:
  • Approaches to ground (multi-modal) generation on up-to-date world, domain-specific, enterprise, or personal knowledge
  • Memory-augmented generative AI systems
  • AI truthfulness and hallucination reduction
  • Agentic AI:
  • Creation of autonomous systems capable of performing tasks, making decisions, and interacting with their environments
  • Multi-Agent systems and agent orchestration framework improvements
  • Customization and continual improvement of agents post deployment
  • Internationalization agentic systems
  • Responsible Generative AI:
  • Red teaming (e.g., advancing red teaming approaches, automated red teaming for multi-modal models)
  • Improvement of foundation model RAI performance (e.g., robustness to jailbreaking and membership inference attacks; watermarking approaches; deepfake detection)
  • Responsible agentic AI (e.g., robustness of multi-agent systems, adherence to guardrails)
  • Privacy-preserving Generative AI
  • Interpretable AI:
  • Understanding mechanisms behind emergent AI capabilities (e.g. mechanistic interpretation)
  • Communicating AI agents’ rationales and search process (meta-cognition) with human stakeholders
  • Small-size LLM:
  • LLM distillation
  • LLM customization for specific skill
  • Knowledge-graph aware LLM
  • Long-context, small-size LLM customized for reasoning capability
  • Applications of Generative AI:
  • Systems that leverage Generative AI for advancing science and technology in areas such as physics, mathematics, chemistry, biology, hardware design, materials science, engineering, economics, healthcare, climate
  • Security:
  • Low-cost computer vision techniques for object interactions for physical security
  • Near-real-time anomaly detection at scale (e.g., streaming petabytes per hour) to identify malicious activity in Linux cloud-based environments
  • Running distributed big data, AI/ML, and/or streaming models with minimal latency and cost using host, network, and audit telemetry sources. Such data engineering can be used for selecting interesting subsets of data including complex, multi-event sequences, improving matching and micro-summarization efficiencies for continuous log processing, and improving event summarization processing efficiency while maintaining accuracy
  • Use of AI/ML (e.g., GenAI) to automate and improve security response workflows (for example, from written reports or threat detections), specifically through incident summarization, response planning, and communication management

Eligibility

Full-time tenure-track and research-track faculty members with a primary appointment in the Whiting School of Engineering are eligible to submit proposals as principal investigators. Collaborative proposals led by WSE faculty with faculty from other JHU divisions are also welcome.

Note: Faculty who will be Amazon Scholars in AY 2025-2026 are eligible to submit proposals, but must adhere to  JHU Conflict of Interest policies and procedures. These individuals are encouraged to consult with Laura Evans at [email protected] well in advance of proposal submission.

Award Types

Two types of awards are available:

  1. Sponsored research awards, with terms comparable to those for other sponsored projects, are expected to involve direct collaboration with Amazon researchers. A few (1-3) awards with up to $100,000 in direct costs are expected to be given in AY 2025-2026.
  2. Gift-funded awards, are treated as unrestricted gifts and are expected to support research that is more exploratory in nature. Several (2-4) awards of up to $100,000 each are expected to be given in AY 2025-2026.

The application process for both types of awards is the same.  The award type will be decided in consultation with Amazon.  Applicable indirect charges will be added to the awarded proposals.

Important Dates

  • February 28, 2025: Pre-proposals/Abstracts due (PDF, 1-page max, via the AI2AI portal)
    Optional, but strongly encouraged – pre-proposals will receive feedback from Amazon scientists to strengthen proposal
  • March 31, 2025: Full proposals due
  • Summer 2025: Award decisions made
  • The award period is nominally September 1, 2025 to August 31, 2026.

Format

Pre-proposal – due February 28th (optional but strongly encouraged)

  1. Proposal abstract/brief overview (PDF, 1-page maximum)
    1. Technical description of the project
    2. Expected deliverables/outcomes, and milestones
Submit Pre-proposal

Due by February 28th

Full Proposal – due March 31st

  1. Proposal overview (PDF, 2-4 pages)
    1. Technical description of the project
    2. Expected deliverables/outcomes, and milestones
  2. Budget and justification (Excel, using the provided template)
  3. References (unlimited)
  4. Biographies of the PI’s (up to 3 pages per PI in NSF Format)

Proposals currently under consideration by the Amazon Research Awards program should be noted to indicate concurrent submissions.  Proposals will not be selected for funding from both sources.

Submit Full Proposal

Due by March 31st

Funding Guidance and Privacy Policy

Funding Decisions
Amazon may award projects under this CFP as either Gift-Funded Research or Sponsored Research. Projects will typically be funded at the level of up to $100,000 in direct costs, depending on the project scope and nature of funding. If a project is pursued as Sponsored Research, the AI2AI Advisory Board will discuss this adjustment in conjunction with the proposer, and ensure that applicable indirect costs are covered for awarded projects. Sponsored Research Project recipients will not be asked to reduce their budgets to accommodate indirect costs.

Amazon has discretion to assess submitted proposals according to its own criteria that it deems relevant to the evaluation process (e.g., potential impact, suitability of techniques used to address the problem, and feasibility of completing the project within the one-year timeframe). All award decisions relating to this CFP will be final.

Application Content
No proposal to this CFP may contain any confidential information and no part may be marked as ‘confidential.’ Amazon does not accept any legal obligation (whether of confidentiality, compensation, return or otherwise) with respect to any proposals. Amazon may use, edit, modify, copy, reproduce, and distribute all or a portion of the proposal within Amazon’s organization for the purpose of managing the AI2AI Center, including for evaluating the contents of submitted proposals and for matching interested Amazon scientists and teams to funded proposals and university research members. Amazon reserves the right to implement competitive, similar, or identical ideas in the future, without restriction or obligation. You understand and acknowledge that Amazon has wide access to technology, designs, and other materials, and may work on and/or develop projects and ideas that may be competitive with, similar to, or identical to your proposal in theme, idea, format or other respects, inclusive. You acknowledge and agree that you will not be entitled to any compensation as a result of Amazon’s use of any such similar or identical material that has or may come to Amazon from other sources.

When preparing your proposal, please focus on research objectives that can be achieved without access to nonpublic Amazon data or data with noncommercial license restrictions. Proposals should be scoped to use publicly available information or data that can be independently collected or generated, as Amazon does not plan to share any proprietary or confidential data for this call.

Privacy
You acknowledge and agree that we may collect, store, share, and otherwise use personally identifiable information provided during this CFP, including but not limited to, name, mailing address, phone number, and email address. All personally identifiable information collected is subject to, and will be used in accordance with, the Amazon Privacy Notice, including for administering the CFP and verifying applicant’s identities, addresses, and telephone numbers in the event a proposal is selected for funding. By participating in the CFP, you consent to the transfer of personal data to the United States for purposes of administering the CFP and additional purposes that are consistent with goals relating to the USC Center. The data controller for information collected by us is Amazon.com Services, Inc., 410 Terry Ave North, Seattle, Washington 98109, USA.