We are pleased to open the call for proposals for hosting challenge problems for the WSDM Cup 2026, a high impact ML / AI competition event that will be co-located with the WSDM 2026 main conference in Boise, Idaho, USA from February 22-26, 2026.
Competitions of this form are a crucial part of the research ecosystem, bringing together world-class experts from all over the world to independently evaluate their own best ideas on an important problem. Past experience has shown this process to create both amazing progress on specific problems, but also help to achieve a depth of validation and understanding in a way that lives up to the highest ideals of empirical rigor for AI and ML as research fields.
We are seeking proposals for challenge problems and datasets from both industrial and academic institutions, with an emphasis on novel challenges that will help uncover the most topical and important issues facing the WSDM community today.
We are particularly interested in, but do not limit our attention to tasks that can leverage new insights, tasks, and solutions that are aided by the use of Generative AI. Much like many similar shared tasks, we view these types of challenges as competitions that allow academic and industry practitioners to come together to explore interesting problems and discuss successes, challenges, and future directions. Past WSDM Cup tasks have helped provide new datasets and evaluation methodologies, and have introduced tasks to wider audiences that may have been unfamiliar with them.
We strongly encourage all interested institutions and research groups to submit proposals and be part of this important challenge. Hosting the challenge is a fantastic way to bring a strong spotlight to previously under-studied areas or emerging topics, to showcase expertise, to establish new benchmarks in the field, and to contribute to advancing the field forward on real world problems. Moreover, this year the winning proposal for the WSDM Cup Challenge will be an archival submission to be included in a companion volume for WSDM workshops.
Important Dates
- Oct 13, 2025: Proposals due
- Oct 20, 2025: Proposal acceptance notifications
- Oct. 20 – Nov. 16: Cup preparation
- Nov. 17, 2025: Cup starts
- Feb. 6, 2026: Cup ends
- Feb. 22-26, 2026: Winners announced
Requirements
Submissions will be reviewed on their ability to meet the following requirements:
- A clear, well justified description of the problem, including why it is broadly important for the WSDM community in terms of novelty, significance, or timeliness.
- A challenging but manageable and well scoped problem. In particular, there should be a reasonable assessment that the problem is both feasible to approach and that it has appropriate “headroom” for remaining advances and discoveries to make improvements.
- An ideal challenge would present a new challenge that has arisen due to the increasing prevalence of generative systems or a twist on an existing problem area that was previously infeasible.
- A concise and well defined description of any data that will be provided. This should include notes on timelines for collection, which should be completed and allow time for verification and checking for issues such as leakage well in advance of the competitions start date. Additionally, any issues around data licensing should be described. Lastly, it is important that any private test data be truly private, and have never appeared previously in any published or internet-accessible public form.
- Accessibility for non-domain experts. The proposal should describe how researchers who are generally familiar with AI and ML but perhaps not familiar with the specific problem domain may engage with the problem, or what additional materials will be provided to help give appropriate context.
- In particular, domain-specific data and tasks are fine so long as in-depth domain knowledge is not required to analyze failure cases. For example, summarization of medical records or contracts could form the basis of a challenge but a task involving assigning medical billing codes or complex contract understanding would not.
- A clear description of evaluation metrics and evaluation protocols. In order to ensure objective and fair evaluations, and to ensure strong research value as an outcome of the challenge, it is critical that metrics and evaluation protocols are carefully thought through and described in exact detail. Automated evaluation is preferred wherever possible due to cost, scale, and repeatability.
- A complete timeline of the competition, including timeline for data collection, preparation, and validation, for the competition itself to run, and for any post-competition verification, and for knowledge sharing following the close of the competition.
Please submit your proposals via EasyChair:
https://easychair.org/conferences/?conf=wsdm2026 (select WSDM Cup Track). In using EasyChair, your submission will form an archival submission to a companion volume containing WSDM 2026 workshops to further awareness of your efforts.