UNICORN Banner
Join
By joining and submitting to the UNICORN Challenge, participants confirm that they have read, understood, and agree to comply with all rules and guidelines outlined herein, in addition to the Grand Challenge terms and policies. Participants and organizers adhere to all policies.

General guidelines 📜

User Accounts and Team Formation

  • Each participant is required to sign up with one verified account.
  • Anonymous participation is not allowed, only verified Grand Challenge accounts are accepted to participate. Participants must accurately display their true names and affiliations (university, institute, or company), and country of residence.
  • All participants must form a team, even if the team consists of a single member.
  • Each participant can only be a member of a single team.
  • Members from the same institution or group should form a single team.
  • Teams are not allowed to split.
  • Submission limits are defined at the level of a team.

Conflict of interest Individuals directly involved in organizing the challenge, companies sponsoring the challenge, and individuals providing data for evaluation are not eligible to receive prizes. Teams that include such individuals may still participate and be ranked, but their involvement must be clearly disclosed in the model description required for submission to the test phase and visibly indicated on the public leaderboard.

Disqualification The UNICORN challenge organizers reserve the right to disqualify any participant or team at any time for unfair or dishonest practices.

Withdrawal policy Participants reserve the right to withdraw from the UNICORN challenge and forego any further participation. However, prior submissions cannot be retracted, and published results will remain accessible.

Submission components By submitting to UNICORN, participants are aware that each submission consists of two components: the Model and the Adaptor (see Submissions).

Model

Model ownership  The submitted model, including its weights, remains the participant’s property.

Model privacy  Participants are not obliged to release the submitted model publicly, as it may remain private.

Model Usage and Continued Evaluation Models submitted to Grand Challenge will only be executed for UNICORN challenge evaluation purposes, both during and after the challenge for additional validation on relevant datasets. By submitting to the original UNICORN challenge, participants agree to also participate in the post-challenge evaluation. This requires re-submitting the same model used during the test phase to additional data, potentially with modifications to the input interface to support new data formats or tasks. The model's internal logic or learned parameters must remain unchanged. The results will be solely used to contribute to a scientific publication, for which participants will be invited as co-authors. No long-term access to submitted algorithms is required.

Use Restrictions UNICORN organizers will not have direct access to model weights. All model executions will occur through the Grand Challenge platform. Submitted algorithms will not be used for unrelated research, demonstrations, publications, or benchmarking outside of the scope of the UNICORN challenge. UNICORN organizers will not license, sell, rent, host as a service, or otherwise use the algorithm image, model weights, or any derivatives for commercial, regulatory, or revenue-generating purposes.

Adaptor

Adaptor Submission The UNICORN challenge makes a set of adaptors available. Participants can develop new (custom) adaptors and use those in the challenge. Participants who wish to use a custom adaptor must submit it to the official UNICORN adaptor repository. Note that only adaptors from this repository, either provided by the organizers or submitted by participants, can be used for model evaluation on Grand Challenge. Participants can submit custom adaptors through a pull request. Once checked by organizers, these adaptors will be merged into the adaptor repository.

Adaptor privacy In addition to submitting it to the official UNICORN adaptor repository, if participants use a custom adaptor and wish to be eligible for prize money, they commit to making the source code of the adaptor open source. Even if not made open source, participants commit to disclosing the full adaptor implementation to organizers, to be presented at the UNICORN MICCAI workshop and in scientific publications.

UNICORN Docker guidelines 🐋

  • Only fully automated Docker containers are supported. Semi-automated or interactive methods are not allowed.
  • Each team must comply with the compute limits and budget allocation. The algorithm Docker image must not exceed 20 GB in size. Details on the time limits for each task are available here.
  • All submitted Docker containers will run in an offline environment without internet access. Any necessary resources (e.g., pre-trained weights, packages) must be included in the container before submission.
  • Pre-trained models are allowed for foundation model training.
  • Pre-trained adaptation (i.e., in the adaptor) methods are not allowed.
  • Only the few-shots provided to the evaluation Docker may be used to train an adaptation method in the adaptor.
  • Custom adaptation methods must be uploaded to the UNICORN Github repository and will be public.
  • Foundation model weights/feature extractors are not made public, except for prize-winning teams.

Final test phase and prizes 🏆

The UNICORN challenge offers prizes for top-performing teams. To submit to the final test phase, it is required to submit a detailed description in the form of a short scientific report (approximately 2-3 pages) of the training data and methodology, including details about:

  • The datasets used during training, including quantity.
  • Whether the data is publicly available, and if so, where they can be found.
  • Model architecture used during training and number of parameters the model has.
  • Training procedure and resources, including training time, cloud provider/ data center was used and where is it geographically located.
  • For language tasks, if a system prompt is used, it must also be accessible to the organizers.

Open source To be eligible for prize money, participants commit to making the source code of the adaptor and their foundation model weights open source.

Prize money restrictions The competition is open to participants worldwide, except if you are a resident of North Korea, Crimea or any other place covered by Financial measures, Restrictions on goods EU sanctions. The aforementioned regulation is in place because of a requirement to conform to EU regulations concerning the provision of resources and transfer of finances (challenge award money, etc.).

Task coverage Participation in all tasks in the UNICORN challenge is not mandatory. However, for the final leaderboard, any missing task will receive a score of 0 or the minimum possible value for the relevant evaluation metric.