About this CFP
Amazon's Devices and Services organization delivers a constellation of devices, services, and experiences that empower customers to stay connected, safe, informed, productive, educated, and entertained. The portfolio includes Echo, Alexa+, Fire TV, Fire tablets, Kindle, Ring, Blink, and eero. Our science and engineering teams work across audio, camera, and sensor systems, and on the on-device intelligence that turns what a device measures into something useful to the customer.
Models that understand everyday human activity need large, diverse, densely annotated multimodal training data, which is expensive, slow, and privacy-sensitive to collect from real environments. Synthetic generation is the practical alternative, as producing more data is cheap. What remains hard is broader activity coverage, causal consistency across modalities, longer coherent episodes, and control over which scenario gets generated. This CFP solicits research that makes each of these four scale with compute rather than hours of human authoring.
We welcome proposals in the following research topics:
Research topics
Proposals may address more than one topic.
1. Text-to-simulation motion and interaction synthesis. Generating physically plausible human motion and object interaction from natural-language description, conditioned on scene layout and object arrangement. Of interest:
- Motion that respects scene geometry and contact constraints
- Dexterous hand-object interaction rather than coarse body pose
The open problem is moving text-to-motion beyond isolated, abstract poses toward scene-conditioned, contact-accurate sequences.
2. Causally coherent audio-visual synthesis. Generating audio, speech and video that are causally consistent rather than merely synchronized, so that the character of a sound follows from the visible physical event that produced it and spatial audio reflects the geometry of the rendered scene. We are interested in hybrids that combine physics-parameterized sound generation, which is grounded but perceptually limited, with generative audio, which is perceptually convincing but physically unconstrained. Evaluation methodology is as important as generation: metrics for cross-modal causal coherence are largely absent today.
3. Long-horizon multimodal episode generation. Generating hours-scale behavioral episodes from first-person or fixed-viewpoint capture, such as routines, extended tasks, and transitions between activities. World state should persist across the full span. Of interest:
- Propagating accumulated scene state through multimodal sensor streams, so that a change made early in an episode still holds later rather than being discarded when the next segment begins
- Modeling viewpoint and attention, and how sensor characteristics change with mounting position
- Maintaining coherence over horizons far longer than the seconds-to-minutes range typical of video generation today
4. Action-conditioned scenario synthesis with world models. Learning a unified structured representation of environment state that can be rolled forward under specified agent actions, so that the same initial state produces distinct, plausible outcomes under different actions. We are interested in approaches that close the gap between simulation and generative methods, retaining multimodality and long horizons while achieving generative fidelity and generalization.
Theoretical advances, creative new ideas, and practical applications are all welcome.
Timeline
Submission period: October 1 — November 4, 2026 (11:59PM Pacific Time).
Decision letters will be sent out in February 2027.
Award details
Selected Principal Investigators (PIs) may receive the following:
- Unrestricted funds, no more than $80,000 USD on average
- AWS Promotional Credits, no more than $40,000 USD on average
- Training resources, including AWS tutorials and hands-on sessions with Amazon scientists and engineers
Awards are structured as one-time unrestricted gifts. The budget should include a list of expected costs specified in USD, and should not include administrative overhead costs. The final award amount will be determined by the awards panel.
Eligibility requirements
Please refer to the ARA Program rules on the Rules and Eligibility page.
Proposal requirements
Proposals should be prepared according to the ARA proposal template and are encouraged to be a maximum of 4 pages, not including Appendices. In addition, to submit a proposal for this CFP, please also include the following information:
- Which of the research topics above your proposal targets
- How your approach differs from or builds upon existing methods in the area
- How you will evaluate the work, including the metrics or benchmarks used, or the evaluation you propose to establish where none exist
- What will be achieved within the one-year award period
- How the generated data or models would be used in practice, including the formats or interfaces you would produce
- Please list the open-source code, datasets, or benchmarks you plan to contribute
- Please list the AWS tools and services you will use
Selection criteria
Proposals will be reviewed by a panel of Amazon scientists and engineers. Proposals will be evaluated on the following:
- Creativity and quality of the scientific content
- Fit to one or more of the research topics above
- Feasibility of the proposed approach within the one-year award period
- Rigor of the proposed evaluation, particularly where standard benchmarks do not yet exist
- Interest expressed in open-sourcing code, datasets, and benchmarks
Expectations from recipients
To the extent deemed reasonable, award recipients may acknowledge support from ARA (e.g., (“Research reported in this [publication/press release] was supported by an Amazon Research Award, [Cycle /Year].“). Award recipients will inform ARA of publications, presentations, code and data releases, blogs/social media posts, and other speaking engagements referencing the results of the supported research or the Award. Award recipients are expected to provide updates and feedback to ARA via surveys or reports on the status of their research. Award recipients will have an opportunity to work with ARA on an informational statement about the awarded project that may be used to generate visibility for their institutions and ARA.
