The 1^st EReL@MIR Workshop on Efficient Representation Learning for
Multimodal Information Retrieval

Multimodal representation learning has garnered significant attention in the AI community, largely due to the success of large pre-trained multimodal foundation models like LLaMA, GPT, Mistral, and CLIP. These models have achieved remarkable performance across various tasks of multimodal information retrieval (MIR), including web search, cross-modal retrieval, and recommender systems, etc. However, due to their enormous parameter sizes, significant efficiency challenges emerge across training, deployment, and inference stages when adapting these models’ representation for IR tasks. These challenges present substantial obstacles to the practical adaptation of foundation models for representation learning in information retrieval tasks.

To address these pressing issues, we propose organizing the first EReL@MIR workshop at the Web Conference 2025, inviting participants to explore novel solutions, emerging problems, challenges, efficiency evaluation metrics and benchmarks. This workshop aims to provide a platform for both academic and industry researchers to engage in discussions, share insights, and foster collaboration toward achieving efficient and effective representation learning for multimodal information retrieval in the era of large foundation models.

Call for Challenge Submissions (TBA)

We are excited to announce the MIR Challenge featuring two major themes: the Multimodal Document Retrieval Challenge and the MM-CTR Challenge, each with two tracks focused on advancing multimodal retrieval and recommendation systems. We warmly invite researchers and practitioners to participate and showcase innovative solutions in this rapidly evolving field. Details will be announced soon at the following pages: MIR Challenge

Call for Papers

We invite researchers to submit their latest work to the EReL@MIR Workshop on fundamental challenges in multimodal representation learning for Multimodal Information Retrieval (MIR). The topics of interest include, but are not limited to:

Efficient Multimodal Representation Adaptation based on Multimodal Foundation Models
Data-Efficiency in Multimodal Representation Learning
Efficient Multimodal Fusion for Representation Learning
Efficient Cross-Modality Interaction for MIR
Real-Time Inference for Multimodal Representations
Efficient MIR Foundation Models
Benchmarks and Metrics for Multimodal Representation Learning Efficiency

Submission Guidelines

Submissions of papers must be at least 4 pages and at most 8 pages (including figures, tables, proofs, appendixes, acknowledgments, and any content except references) in length, with unlimited pages for references. Submissions of papers must be in English, in PDF format, in the current ACM two-column conference format. Suitable LaTeX, Word, and Overleaf templates are available from the ACM Website (use the “sigconf” proceedings template for LaTeX and the Interim Template for Word).

The review process of the submitted manuscripts will be done together with our program committee. The selection will depend on the technical soundness and relevance of submissions to the community that the workshop is targeting. At least one author of each accepted paper must attend the workshop on-site and present their work. Submissions must be anonymous and should be submitted electronically via EasyChair: https://easychair.org/conferences/?conf=erelmir2025.

WWW2025 Fast Track (new)

WWW2025 Fast Track - Papers rejected and/or withdrawn from WWW 2025 that wish to be submitted to EReL@MIR can use the same submission link. These submissions should include the review comments in an appendix. Such papers will bypass the peer-review process, with acceptance decisions made directly by the meta-reviewers.

In the Appendix, the authors are encouraged to include a section titled “Improvements,” where they briefly outline whether they have addressed any of the issues raised in the main review and, if so, how these revisions were implemented. Only minor revisions are expected, and it is acceptable if no revisions are made where the work is already deemed to be in good shape. Major changes are not permitted, and any attempt to modify or manipulate review comments is strictly prohibited.

Acceptance and the Best Paper Award

Authors of accepted papers may choose whether to include their work in the WWW’25 Companion proceedings. We will reach out to the authors of accepted papers at a later time to facilitate this decision. The committee will review the submissions and select one outstanding workshop paper to receive the Best Paper Award at the EReL@MIR Workshop.