Explore

IR-RAG @ SIGIR25

Information Retrieval's Role in RAG Systems

Find the schedule

here⁠

⁠

Find the important dates

here⁠

⁠

Find the accepted papers

here⁠

⁠

If you are a RAG-enthusiast, this is the workshop for you

Overview

In recent years, Retrieval-Augmented Generation (RAG) systems have become a cornerstone of artificial intelligence, attracting considerable attention in a variety of fields. By integrating the strengths of information retrieval and generative models, these systems have shown immense potential to push the boundaries of machine learning applications. Nevertheless, RAG systems still face significant challenges and offer ample room for advancement and innovation.

This workshop aims to highlight the central role of information retrieval within RAG frameworks, which we believe has often been overshadowed by the emphasis on the generative components. While the generative models are integral to these systems, the quality and effectiveness of the retrieval mechanism is equally critical, as it has a direct impact on the overall system performance and outcomes.

We invite papers that rethink and prioritise the fundamental aspects of RAG systems, particularly in strengthening the information retrieval component. Through this workshop, we aim to gain deeper insights into how improved retrieval methods can enhance the performance and reliability of RAG systems.

The event will bring together leading experts, researchers and practitioners to provide a collaborative platform for exchanging ideas, sharing results and fostering innovation. Our aim is to stimulate research and discussion that reaffirms the essential role of information retrieval in shaping the next generation of generative systems.

Important Dates

Submission deadline: April 23, 2025

Acceptance Notification: May 21, 2025

IR-RAG Workshop: July 17, 2025

Camera-ready versions of accepted papers due: TBD (after the conference)

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

Call for Papers

The primary goal of this workshop is to bring much-needed attention to the retrieval mechanism within RAG systems, while exploring the central question:

How should information retrieval research evolve in the era of RAG systems?

By bringing together experts, practitioners and enthusiasts, the workshop aims to foster dialogue and innovation around the challenges and opportunities associated with the retrieval component of RAG systems. The goal is to cultivate a robust community dedicated to advancing this critical aspect of RAG, ensuring that it receives equal focus alongside the generative elements. Through this platform, we aim to share knowledge, inspire new research directions, and foster collaborative efforts to better understand and improve retrieval mechanisms. Ultimately, we envision building a strong foundation for a thriving community committed to shaping the future of information retrieval for RAG systems. We invite papers that explore a variety of topics, including but not limited to

Use Of The Retrieved Context By The LLM:

Studies \cite{lewis2020retrieval,sauchuk2022role,cuconasu2024power} have shown that RAG systems are sensitive to type and order of retrieved contexts. These findings open avenues for deeper exploration of how retrieved information influences generative output.

(Query) Representation Learning:

Improving queries' representation can significantly enhance a retriever's ability to find relevant documents. This includes using advanced NLP techniques to better capture the contextual subtleties and intent behind queries.

Incorporating Contextual Information:

Incorporating broader contextual elements (e.g. conversation history, user preferences) into the search process can lead to more accurate results.

Updating the Document Database:

Research into efficient update mechanisms and real-time synchronisation. which ensures that document repositories are up to date, is encouraged.

Reducing Computational Load:

Optimising retrieval methods for speed and efficiency, especially on large datasets, can make these systems more practical for real-time applications.

Bias Mitigation:

Investigating ways to identify and reduce bias in retrieval processes is crucial to ensure the fairness and reliability of the content generated.

Cross-Lingual Retrieval Capabilities:

Improving retrieval capabilities in multilingual environments is essential for systems serving diverse audiences.

Multimodality:

While most research has focused on text-based RAG systems, multimodal capabilities are increasingly important for a wide range of applications. Research into methods for incorporating and retrieving data from multiple modalities is encouraged.

Other:

We welcome novel contributions that present unexplored challenges or propose innovative ideas to advance the field and further enrich the discourse.

Submission Instructions

All submissions will be peer reviewed (double-blind) by the program committee and judged by their relevance to the workshop, especially to the main themes identified above, and their potential to generate discussion. All submission must be written in English and formatted according to the latest ACM SIG proceedings template available at http://www.acm.org/publications/proceedings-template.

Submissions must describe work that is not previously published, not accepted for publication elsewhere, and not currently under review elsewhere.

The workshop follows a double-blind reviewing process. Please note that at least one of the authors of each accepted paper must register for the workshop and present the paper.

We invite research contributions, position, demo and opinion papers. Submissions must either be short (at most 4 pages) or full papers (at most 9 pages). References do not count against the page limit.

We encourage but do not require authors to release any code and/or datasets associated with their paper.

Submission website:

https://easychair.org/conferences/?conf=irragsigir24⁠

⁠