Call for Contributions

Call for Full Papers

The 19th ACM Conference on Recommender Systems (RecSys 2025), the leading conference for research on the foundations and applications of recommendation technologies, will take place from September 22 to 26, 2025, in Prague, Czech Republic.

We look forward to receiving your contributions to the FULL research paper track for RecSys 2025. Below, you will find the “In-Brief” and “Important Dates” sections of the CFP, which discuss key points of attention for this call. In the rest of the CFP, we provide detailed information that authors should thoroughly review when preparing their submissions.

For the full paper track, we welcome high-impact original papers that contribute to all aspects of recommender systems. CFPs for other RecSys 2025 tracks will be released separately:

  • Short papers: contributions that typically discuss exciting work that is not yet mature enough for the full paper track, for example, due to limited experimental validation.
  • Demo: implementations of novel, interesting, and important recommender systems’ concepts or applications.
  • Reproducibility: contributions that discuss several aspects of reproducibility of empirical results, such as new resources or novel evaluation methodologies.
  • Industry: papers that discuss field experiences, deployments, user studies and real-world challenges faced by industry practitioners.
  • Late-breaking results: short presentations of preliminary work, mainly focused on fostering discussions with other members of the RecSys community.

IN BRIEF

The RecSys community values recommender systems’ human and economic impacts as much as the underlying algorithms. Therefore, we invite contributions to RecSys 2025 that cover the *full spectrum* of recommender system research. As always, we will prioritize the quality of the submitted work; however, this year, we will place even greater emphasis on the impact of each contribution. “A Guide to Writing the NeurIPS Impact Statement” provides non-binding guidance on some of the elements authors may wish to consider in assessing the impact of their work.

  • Submissions will be handled electronically via EasyChair. The abstract is due by April 1st, 2025 (required), and the full paper is expected by April 8th, 2025.
  • Submissions should be anonymous and will be reviewed by at least three members of the Program Committee and overseen by a Senior PC member. There will be a rebuttal phase during which authors can provide a brief narrative to clarify any misconceptions that may have arisen from the reviews and to address specific issues for which an answer is critical to the assessment of the paper.
  • At the time of submission, information for all authors and authorship order must be established — no changes will be allowed after the review process begins. Moreover, authors must declare conflicts of interest with PC and SPC members before submitting the full paper. Failure to declare COI can result in desk-rejection.
  • RecSys 2025 has a no dual submission policy, which means submitted manuscripts must not be under review at another publication venue, including other RecSys 2025 tracks. Submission of the paper to pre-print sites like arXiv is allowed as long as the title and abstract of your paper in the archival platform are distinct from your RecSys submission.
  • Contributions must be self-contained, limited to 8 pages (using the ACM 2-column template), with up to 2 pages for references. Note that appendices also count towards the 8-page limit. No supplementary materials will be accepted; instead, we encourage authors to include a link to an anonymous repository for code, pre-registered studies, and dataset information within their paper.
  • Contributions will be reviewed based on their relevance to the RecSys community, scientific rigor, and impact.
  • At least one author from each accepted submission must attend the conference in person to present their work and participate in the Q&A session. Remote presentations or videos will not be permitted for accepted contributions.

IMPORTANT DATES FOR FULL PAPERS

Deadlines refer to 23:59 (11:59pm) in the AoE (Anywhere on Earth) time zone.

  • Full paper abstracts due: April 1, 2025
  • Full papers due: April 8, 2025
  • Author rebuttal period for Full papers: May 20-26, 2025
  • Full paper notifications: July 3, 2025
  • Camera-ready due: July 21, 2025 (strict)

IN-PERSON ATTENDANCE POLICY

RecSys 2025 is an in-person conference. All accepted papers to the main conference are expected to be presented in-person. At least one author from each accepted paper must attend the conference in person to present the paper and address audience questions during the Q&A session. No pre-recorded videos will be permitted for Full papers. Papers that are not presented at the conference by an author may be removed from the proceedings at the discretion of the program chairs.

SUBMISSION GUIDELINES

Submitted contributions should focus on original and significant work that provides lasting value. The research should address the theory and/or practice of recommender systems. We welcome contributions that showcase innovative uses of algorithms, novel interfaces and explore the benefits and challenges of applying recommender system technology in real-world applications that impact a broad audience and environment.

Evaluations of proposed solutions or applications must align with the claims made in the paper. Depending on the nature of the contribution, this could include simulation studies, offline evaluations, A/B tests, or controlled user experiments.
Research methods and technical procedures should be detailed enough to allow for scrutiny and reproducibility. In other words, contributions should be self-contained. While we understand that user data may be proprietary or confidential, we encourage the sharing of anonymized and cleaned datasets, along with data collection procedures and code. Hence, to enhance reproducibility, we encourage authors to include in their manuscript, when pertinent, a link to an anonymous repository where they can include for example their code or data. Authors can also submit auxiliary material to enhance reproducibility (e.g., experiment details, proof details, appendixes, and study pre-registration). For this, the authors may refer to https://anonymous.4open.science/.

Results should be communicated clearly, and the implications of the findings for RecSys and other fields should be explicitly discussed.

Full research papers must describe original work that has not been previously published, not accepted for publication elsewhere, and not simultaneously submitted or currently under review in another journal or conference, including the other tracks of RecSys 2025. See this brief checklist to strengthen a RecSys paper, for authors and reviewers.

Submissions of full research papers must be in English and be at most 8 pages in length including figures, tables, proofs, appendixes, acknowledgements, and any content except references, plus up to 2 pages of references.
Each accepted long paper will be included in the conference proceedings and presented in the form of a talk or poster during the main conference program.

Submissions must be anonymous and should be submitted as a PDF electronically via EasyChair:
link by selecting the “RecSys 2025 Full Papers” track (link will be provided later).

The review of manuscripts will be mutually anonymous, and submissions not properly anonymized will be desk-rejected without review.

RELEVANT AREAS & TOPICS

  • Human-centered recommendation approaches
    • Adaptive algorithms that adjust recommendations dynamically based on user interactions
    • Explanation methods and interfaces for recommender systems
    • Human-in-the-loop model learning and validation
    • Innovative user interfaces for recommender systems
    • Novel interaction paradigms
    • Novel methods for improving the explainability of recommendation models
    • Novel methods for preference elicitation
    • Novel perspectives on the role of transparency in recommender systems
    • Studies that investigate how real-time events (e.g., trends and holidays) influence user preferences
    • User control of RecSys
    • User experiments and studies of recommendation applications
  • Societal recommender systems
    • Adapting recommendation algorithms to suit different cultural contexts and multilingual users
    • Addressing global digital divides by exploring the design, development and deployment of systems for low-resource environments
    • Bias, fairness, bubbles, and ethics of recommender systems
    • Diversity and inclusion models
    • Eco-aware recommendation models
    • Ensuring equitable access to digital resources for understudied populations
    • Implications of recommendation algorithms when the main stakeholders are understudied populations
    • Recommendation models for sustainable tourism
    • Studying the effects of recommendation systems on cultural and global diversity lenses
    • Supporting understudied communities by elevating diverse and underrepresented content
    • Sustainable Recommender Systems Development
  • Computational innovations, evaluation, and real-world applications
    • Building large-scale, standardized datasets for benchmarking algorithms
    • Case studies of real-world implementations
    • Conversational and natural language recommender systems
    • Cross-domain recommendation
    • Design of usability studies
    • Economic models and consequences of recommender systems
    • Generative models in recommender systems
    • Legal and regulatory aspects of recommender systems
    • Knowledge-based recommender systems
    • Multimodal approaches for recommendation
    • Multi-stakeholder recommendations
    • Novel evaluation metrics beyond accuracy
    • Privacy and security
    • Recommendation models for education and learning-related technologies
    • Studies that investigate the relationship between evaluation metrics and real-world outcomes
    • Tailoring recommendations for applications beyond e-commerce

FORMATTING

ACM’s archival publication format separates content from presentation in the Digital Library to enhance accessibility and improve the flexibility and resiliency of our publications. All authors should submit manuscripts for review in a double-column format. Instructions for Word and LaTeX authors are given below:

  • Microsoft Word: Write your paper using ACM’s interim template. Follow the embedded instructions to apply the paragraph styles to your various text elements. The text is in double-column format and no additional formatting is required at this stage.
  • LaTeX: Please use the latest version of the Primary Article Template – LaTeX to create your submission.Start the document with the \documentclass[sigconf,anonymous]{acmart} command to generate the output in a double-column format. Please see the LaTeX documentation and ACM’s LaTeX best practices guide for further instructions, ignoring the single-column instructions. Do not use the “manuscript” option, otherwise the document will not be compiled in double-column, as required. Check the sample-sigconf.tex file included in the template package for a formatting example. To ensure 100% compatibility with The ACM Publishing System (TAPS), please restrict the use of packages to the whitelist of approved LaTeX packages.

Further instructions for camera-ready submission will be provided to authors of accepted contributions.

Authors are strongly encouraged to provide “alt text” (alternative text) for floats (images, tables, etc.) in their content so that readers with disabilities can be given descriptive information for these floats that are important to the work. The descriptive text will be displayed in place of a float if the float cannot be loaded. This benefits the author as well as it broadens the reader base for the author’s work. Moreover, the alt text provides in-depth float descriptions to search engine crawlers, which helps to properly index these floats. Please check your template’s documentation for instructions on how to use “alt text”. Additionally, authors should follow the ACM Accessibility Recommendations for Publishing in Color and SIG ACCESS guidelines on describing figures.

Should you have any questions or issues going through the instructions above, please contact support at for both LaTeX and Microsoft Word inquiries.

Accepted papers will be later submitted to ACM’s new production platform where authors will be able to review PDF and HTML output formats before publication.

REVIEW PROCESS

RecSys follows a mutually anonymous review process. Papers will be reviewed by at least three members of the Program Committee and overseen by a Senior PC member, who will assess submissions based on their relevance, originality, rigor, and the impact of their contribution to the field. Reviewers are also asked to consider the replicability of reported research. Papers that are out of scope, incomplete, or lack sufficient evidence to support their main claims may be rejected without a full review.

After the review phase, authors may produce an anonymous rebuttal, whose goal is to refute any factual errors in reviews or to supply additional information or clarifications requested by the reviewers. Rebuttals may include minor additional experiments or analysis requested by reviewers. They may also include figures, graphs or proofs to better illustrate your arguments. Rebuttals MUST NOT add new contributions (theorems, algorithms, experiments) that were absent in the original submission and were not specifically requested by the reviewers. Further information will be provided along with the reviews.

AUTHORSHIP POLICY & CONFLICT of INTERESTS

Authors are strongly encouraged to carefully review ACM’s authorship policy before submitting their papers. All authors must be listed in the correct order in EasyChair by the submission deadline. Moreover, they should adhere to the ACM Conflict of Interest policy. CoI refers to close personal relationships, continuing collaborations in the past 3 years (e.g., co-author on paper, joint grant), past or current advisor/advisee relationship, and employment at the same institution in the past 3 years. For full details, please visit this site: https://www.acm.org/publications/policies/conflict-of-interest

The complete author list must be provided by the abstract submission deadline to assist reviewers in identifying potential conflicts of interest. Therefore, no changes to authorship will be allowed for any reason after the abstract submission deadline, and no updates will be permitted for camera-ready versions.

Please ensure that all authors obtain an ORCID ID, to complete the publishing process for accepted contributions, since as an ACM conference, we have committed to collecting ORCID IDs from all of our published authors. This helps improve author discoverability, ensure proper attribution and contribute to ongoing community efforts around name normalization.

ANONYMITY

The peer review process is mutually anonymous. This means that all submissions must not include any information that identifies the authors or their organizations. Specifically, authors should not include their names and affiliations and must refer to their previous work in the third person. For example, you could write, “de Gemmis and Pera (2025) stated that RecSys submissions should be anonymized, which can be accomplished by referring to the authors’ prior work in the third person.” It is also important to avoid providing any other information that could help reviewers identify the authors, such as acknowledgements of individuals and funding sources.

It is acceptable to explicitly mention the companies or organizations that provided datasets, hosted experiments, or deployed solutions, as long as there is no implication that the authors are currently affiliated with those organizations. Reviewers are instructed not to search for technical reports, pre-prints, or other information about your research. Therefore, it is the authors’ responsibility to ensure that the manuscript submitted does not reveal their identity as the authors.

PRE-PRINT POLICY

Please carefully consider pre-print constraints. Any violation of this policy could cause your paper to be desk-rejected.

  1. An anonymous version of your RecSys submission can be submitted at any time to pre-print sites like arXiv.
  2. Non-anonymized work having significant overlap with your RecSys submission, posted on any online archival platform *before* the submission deadline, must be disclosed in the EasyChair submission form. The title and abstract of the RecSys submission must be distinct from your previous version.
  3. Any non-anonymized version of your RecSys paper cannot be posted to pre-print sites during the review period.

USE of AI POLICY

Authors must follow the ACM Policy on the use of generative AI software tools. If a paper includes material generated by GenAI tools (such as text produced by large language models like ChatGPT), it is essential to disclose the extent and nature of this use in a section titled “Acknowledgments”. This allows reviewers to evaluate the overall rigor of the research methodology. However, the use of AI tools for editing and refining authors’ work—meaning tasks like grammar checks, word autocorrect, and other light editing—does not require disclosure. Authors who choose to utilize GenAI tools are fully responsible for any inaccuracies, biases, plagiarism, and other violations, just as if they had created the content themselves. Submissions that are primarily produced by GenAI without substantial contributions from the authors are prohibited and considered spam. Only humans are permitted to be authors of submitted papers.

PLAGIARISM POLICY

Plagiarism is a matter that as Program Chairs we take very seriously, thus, we check the plagiarism levels of all submitted papers to ensure content originality using an automated tool.
If a contribution reuses non-original text from a previous publication (for example, the description of an algorithm or dataset), please ensure that the prior publication is cited as the source of that text.

For questions regarding reusing text or simultaneous submissions, please contact the program chairs at least one week before the submission deadline. For further details, refer to the ACM Publishing License Agreement and Authorship Policy. Papers violating any of the above guidelines are subject to rejection without review and cases may be referred to the ACM Publications Ethics and Plagiarism committee for further action where warranted.

ETHICAL REVIEW FOR HUMAN-SUBJECTS RESEARCH

ACM RecSys expects all authors to comply with ethical and regulatory guidelines associated with human subjects research, including ACM’s new Publications Policy on Research Involving Human Participants and Subjects. Papers reporting on such human subjects research must include a statement identifying any regulatory review the research is subject to (and identifying the form of approval provided), or explaining the lack of required review. Reviewers will be asked to consider whether the research was conducted in compliance with applicable ethical and regulatory guidelines.

DESK REJECTION POLICY

Submissions that do not meet the requirements for anonymity, length, or formatting; violate dual submission policy and/or any of ACM’s policies on use of AI or academic dishonesty—such as plagiarism, author misrepresentation, or falsification—may be rejected by the chairs without further consideration.
To finalize their submission, authors will be requested to declare conflicts of interest with PC and SPC members. Missing declarations might result in desk rejection.

The ACM Code of Ethics grants Program Chairs the authority to (desk) reject papers that perpetuate harmful stereotypes, employ unethical research practices, or uncritically present outcomes or implications that disadvantage marginalized communities. Additionally, reviewers will be explicitly asked to consider whether the research was conducted per professional ethical standards and applicable regulatory guidelines. Failure to adhere to these standards may result in a rejection. Some concrete examples of violations that may result in desk rejection are:

  • Failures to declare conflicts of interest with PC or SPC members.
  • Any other content after the maximum length of the submission.
  • Wrong template or any attempt to format change to get around the page limit.
  • Authors or their institutional affiliations are explicitly stated or easily discoverable.
  • Links to source code repositories or datasets that reveal, explicitly or implicitly, the identity of one of the authors.
  • Any change to the list of authors (both names and affiliations) after the abstract submission deadline.
  • Double submission to another publication venue.

PUBLICATION and OPEN ACCESS

By submitting a manuscript to an ACM Publication, authors acknowledge that they are subject to all ACM Publications Policies, including ACM’s new Publications Policy on Research Involving Human Participants and Subjects. Alleged violations of this policy or any ACM Publications Policy will be investigated by ACM and may result in a full retraction of the manuscript paper, in addition to other potential penalties, as per ACM Publications Policy.

The official publication date is when the proceedings are made available in the ACM Digital Library. This date may be up to two weeks prior to the first day of the conference.

Authors of accepted papers will have the choice to publish the work either under the conventional ACM policy free of author fees, or under the ACM Open Access policy potentially subject to an article processing charge.

PROGRAM CHAIRS

For CFP-related questions, please reach out to the RecSys 2025 Program Chairs:

  • Marco de Gemmis, University of Bari Aldo Moro, Italy
  • Sole Pera, TU Delft, The Netherlands

Contact information: