ACM Transactions on Information Systems (TOIS) seeks previously unpublished papers on information retrieval (such as search engines, recommender systems, conversational search agents) that contain:

    new principled information retrieval models or algorithms with sound empirical validation;
    observational, experimental and/or theoretical studies yielding new insights into information retrieval or information seeking;
    accounts of applications of existing information retrieval techniques that shed light on the strengths and weaknesses of the techniques;
    formalization of new information retrieval or information seeking tasks and of methods for evaluating the performance on those tasks;
    development of content (text, image, speech, video, etc.) analysis methods to support information retrieval and information seeking;
    development of computational models of user information preferences and interaction behaviors;
    creation and analysis of evaluation methodologies for information retrieval and information seeking; or
    exceptionally well-written surveys of existing work that propose a significant synthesis.

The information retrieval scope of ACM Transactions on Information Systems (TOIS) appeals to industry practitioners for its wealth of creative ideas, and to academic researchers for its descriptions of their colleagues' work.

Special Issue on Causality Representation Learning in LLMs-Driven Recommender Systems
Día de Entrega: 2024-09-30

Guest Editors:
• Lina Yao, CSIRO’s Data61 and University of New South Wales, Australia (lina.yao@unsw.edu.au)
• Julian McAuley, University of California San Diego, United States (jmcauley@eng.ucsd.edu)
• Yongfeng Zhang, Rutgers University, United States (yongfeng.zhang@rutgers.edu)
• Kun Zhang, Carnegie Mellon University, United States, and Mohamed bin Zayed University of
Artificial Intelligence, United Arab Emirates (kunz1@cmu.edu)

The realm of recommender systems has witnessed rapid advancements with the application of deep learning and
other advanced algorithms. While these models excel at predicting user preferences, understanding the
underlying causal relationships has become paramount. The exploration of causal inference in recommender
systems aims to offer transparent, reliable, and more interpretable recommendations.

As the digital realm evolves, recommender systems stand at the crossroads of enhancing user experiences by
providing tailored content. The advent and prominence of Large Language Models (LLMs) open new avenues to
probe deeper into the intricate webs of causality within these systems. LLMs, with their vast knowledge bases and
intricate architectures, present an untapped resource for causal discovery. This call for papers seeks pioneering
research and reviews that explore, analyze, and showcase the synergy between LLMs and causal inference in
recommender systems.

Topics of Interest Include, But Are Not Limited To:
● Foundations of causal inference in recommendation systems.
● Methods for causal effect estimation in user-item interactions.
● Interpretable models based on causal relationships.
● Evaluating the impact of interventions in recommender systems.
● Limitations and challenges in uncovering causalities.
● Trustworthy recommender systems based on causal methods.
● Causal methods for bias and fairness in recommender systems.
● Causal robustness of recommender systems
● Real-world applications and case studies highlighting causal inference benefits.
● Comparison of traditional predictive modelling and causal-based recommendation.
● Theoretical exploration of LLMs in discerning causal relationships within recommender systems.
● Design and development of recommender systems powered by LLM-driven causal insights.
● Leveraging LLMs for enhanced transparency and interpretability in recommendations.
● Addressing inherent biases in LLMs and their impact on causal discovery.
● Case studies demonstrating the practical benefits and challenges of using LLMs for causality in
recommendations.
● Comparative evaluation of LLM-integrated causal discovery against conventional methodologies.
● Novel techniques for training LLMs with a focus on causality.
The proposal for a special section on " Causality Representation Learning in LLMs-Driven Recommender Systems "
within the ACM Transactions on Information Systems is underpinned by a confluence of academic rigor, timely
relevance, and evident momentum within the research community.
● Rapid advancements in machine learning, deep learning, and causality theory have opened new research
avenues that require a specialized forum for dissemination and critique has propelled the field of recommender
systems forward. The complexity and sophistication of modern recommender systems necessitate a nuanced
understanding of the underlying causal mechanisms that inform recommendations, to enhance their
transparency, reliability, and interpretability. This special section aims to serve as such a platform, encouraging
comprehensive exploration and innovation at the intersection of causality and recommender systems.
● The advent of Large Language Models (LLMs) has significantly broadened the scope for investigating causality
within the realm of recommender systems. The expansive knowledge bases and intricate architectures inherent to
LLMs present unprecedented opportunities for causal discovery and analysis, positioning them as a focal point of
contemporary research endeavors.
● The proposed special section is characterized by its inclusivity and breadth, encompassing a wide array of topics
from foundational causal estimation and discovery to the application of LLMs and deep learning techniques in
real-world scenarios. This diversity not only reflects the multifaceted nature of the field but also underscores the
rich potential for scholarly inquiry across various dimensions of causality in recommender systems.
● This call for papers articulates a balanced approach, soliciting contributions that span practical applications and
empirical case studies to theoretical frameworks and methodological explorations. This dual focus is designed to
appeal to a broad spectrum of researchers, encompassing both those engaged in applied research and those
pursuing more theoretical questions within the context of causality in recommender systems.
● This special section is its emphasis on leveraging causal methods to address critical issues of bias and fairness
within recommender systems. This focus is especially pertinent given the escalating concerns surrounding ethical
considerations in AI and machine learning. By prioritizing research that confronts these challenges, the special
section aims to attract submissions that contribute to the development of more equitable and responsible
recommender systems.

The guest editors, affiliated with renowned institutions, bring a wealth of expertise and a broad professional
network to this initiative, further bolstering the likelihood of attracting high-quality submissions. Their previous
organization of related events, such as the Special Issue on Responsible Recommender Systems in the ACM
Transactions on Intelligent Systems and Technology, tutorial on Large Language Models for Recommendation at
RecSys 2023, and workshop on Personalized Generative AI at CIKM 2023, evidences the topic's relevance and the
community's engagement. These preliminary events have not only demonstrated the community's interest but
have also highlighted the depth and quality of research being conducted in this area.

In light of these considerations, the proposed special section on "Causality Representation Learning in LLMs-
Driven Recommender Systems" represents a timely and necessary addition to the scholarly literature. It promises
to consolidate the research community by providing a forum for the exchange of ideas, fostering collaboration,
and establishing a cohesive body of knowledge that advances our understanding of causality in recommender
systems. This proposal stands as a testament to the field's dynamism and the ongoing need for rigorous academic
discourse to navigate its complexities and challenges.

Important Dates:
Submission Deadline: September 30, 2024
First Round Notification: November 30, 2024
First Round Revision: February 28, 2025
Notification of Final Decision: April 30, 2025
Tentative Publication: Mid-to-late 2025

Guest Editors

Lina Yao (https://www.linayao.com/), CSIRO’s Data61 and University of New South Wales, Australia
Lina Yao holds the position of Senior Principal Research Scientist and Science Lead at CSIRO's Data61, alongside
her academic roles as a Conjoint Professor at the University of New South Wales, Honorary Professor at
Macquarie University, and Adjunct Professor at the University of Technology Sydney. Her research endeavors are
deeply rooted in the development of generalizable, transparent, and data-efficient methodologies within the
domains of data mining, machine learning, and deep learning. One of her specialized focuses is deep learning-
based recommender systems, where she strives to address challenges of data scarcity, interoperability, and
complexity of human patterns. Her research delves into understanding the temporal and contextual nuances of
human interactions with digital platforms, enabling the development of recommendation models that perform
effectively and evolve in tandem with user needs and societal trends.

Julian McAuley (https://cseweb.ucsd.edu/~jmcauley/), University of California, San Diego, USA
Julian McAuley has been a professor at UCSD since 2014 (Associate in 2019; Professor in 2021), where his lab
works on problems in the area of Personalized Machine Learning. Broadly speaking, his lab’s research seeks to
develop machine learning techniques for settings where differences among individuals explain significant
variability in outcomes. A core instance of this problem is that of recommender systems, one of the core areas of
his lab’s research, where he develop technologies that underlie algorithms like those used for recommendations
on Netflix, Amazon, or Facebook.

Yongfeng Zhang (https://www.yongfeng.me/), Rutgers University, USA
Yongfeng Zhang is an Assistant Professor in the Department of Computer Science at Rutgers University. His
research interest is in Machine Learning and Data Mining, Information Retrieval and Recommender Systems,
Natural Language Processing, and Trustworthy AI. His research works appear in top-tier computer science
conferences and journals such as SIGIR, WWW, KDD, NeurIPS, ACL, TOIS, TORS, etc. He has been frequently
serving as area chair or senior program committee member at SIGIR, KDD, WWW, RecSys, CIKM, AAAI, etc. He
also serves as Associate Editor for ACM Transactions on Information Systems (TOIS), ACM Transactions on
Recommender Systems (TORS), and Frontiers in Big Data. He is a Siebel Scholar of the class 2015 and an NSF
career awardee in 2021.

Kun Zhang, Carnegie Mellon University & Mohamed bin Zayed University of Artificial Intelligence, USA and UAE
Kun Zhang (http://www.andrew.cmu.edu/user/kunz1/) is currently on leave from Carnegie Mellon University
(CMU), where he is an associate professor of philosophy and an affiliate faculty in the machine learning
department; he is working as a professor of machine learning, the acting chair of the machine learning
department, and the director of the Center for Integrative AI at Mohamed bin Zayed University of Artificial
Intelligence (MBZUAI). He has been actively developing methods for automated causal discovery from various
kinds of data and investigating machine learning problems including transfer learning, concept learning, and deep
learning from a causal perspective. Dr. Zhang co-authored a best student paper for UAI and a best finalist paper
for CVPR, and received an ICML 2022 Test of Time Award Honorable Mention and the best benchmark award of
the causality challenge. He has been frequently serving as a senior area chair, area chair, or senior program
committee member for flagship conferences in machine learning or artificial intelligence, including NeurIPS, ICML,
UAI, IJCAI, AISTATS, and ICLR. He was a co-founder and general and program co-chair of the first Conference on
Causal Learning and Reasoning (CLeaR 2022), a program co-chair of the 38th Conference on Uncertainty in
Artificial Intelligence (UAI 2022), and a general co-chair of UAI 2023.

Special Issue on Query Performance Prediction Towards Novel Information Retrieval Paradigms
Día de Entrega: 2025-03-15

Guest Editors:
• Dr. Suchana Datta, University College Dublin, Ireland, suchana.datta@ucdconnect.ie
• Dr. Guglielmo Faggioli, University of Padua, Italy, guglielmo.faggioli@unipd.it
• Prof. Nicola Ferro, University of Padua, Italy, ferro@dei.unipd.it
• Dr. Debasis Ganguly, University of Glasgow, United Kingdom, debasis.ganguly@glasgow.ac.uk
• Prof. Iadh Ounis, University of Glasgow, United Kingdom, iadh.ounis@glasgow.ac.uk

This Special Issue focuses on Query Performance Prediction (QPP) with respect to recent advances in Information
Retrieval (IR).

QPP is a branch of IR evaluation: it is defined as the task of assessing or predicting the performance of a query
without human-made relevance judgements.

The focus of the special issue will be on three major topics concerning QPP:
• The development of novel QPP models that employ recent neural state-of-the-art solutions, such as
Large Language Models (LLMs) and semantic representations.
• The application of QPP models to novel IR tasks, such as conversational search, fairness-oriented tasks,
multimedia and multimodal retrieval, and Retrieval Augmented Generation (RAG).
• The evaluation of QPP methods performance.

In particular, classical QPP models have been developed focusing on IR systems based on exact matching (i.e.,
systems that consider the presence of a query term in the document as the only relevance signal). Advances in IR,
especially linked to the development of neural IR models and the recent extensive usage of LLMs, have also
highlighted the importance of semantic signals in producing good retrieval results. Therefore, this special issue aims
to attract works, which employ techniques that make use of modern neural IR approaches for the QPP task, thereby
aligning the QPP models to today’s state-of-the-art IR systems.

Secondly, QPP has been traditionally applied to ad-hoc retrieval. In ad-hoc retrieval, the user issues a – typically
short – natural language query and obtains a set of documents in response. Modern IR systems tend to be more
complex with multiple interactions between the user and the system, such as in the case of conversational search,
might include additional modules, such as in the case of RAG, and do not necessarily operate on text, such as in the
case of multimodal or multimodal IR. Therefore, this special issue aims at capturing works that operationalise QPP
for IR in novel domains and tasks.

Finally, evaluating QPP models has always been particularly challenging as we need large amounts of annotated
data and fine-grained evaluation measures. Furthermore, the advent of new IR paradigms, some relying on a
completely different definition of “performance”, has further exacerbated the problem. Thus, this special issue aims
to attract works that propose an advancement over the current state-of-the-art for what concerns QPP evaluation.

Topics
We welcome submissions on the following topics, including but not limited to:
• Application of QPP to Neural Information Retrieval Systems
• Usage of QPP for modern tasks, including, but not limited to, conversational search, fairness, RAG,
multimodal retrieval
• Usage of Large Language Models for QPP
• QPPs based on non-lexical (e.g., semantic, multimodal) signals
• Supervised QPP
• Simulation and construction of evaluation collections with Large Language Models
• QPP evaluation measures
• Development of QPP evaluation collection
• Performance Prediction in neighboring areas including NLP and Recommender Systems
• Theory underneath QPP
• Applications of QPP for downstream tasks, e.g., selective application of second-stage ranking or
relevance feedback.
• Explainability of QPP models and QPP models for explainability

Important Dates
• Submissions deadline: March 15, 2025
• First-round review decisions: May 15, 2025
• Deadline for revision submissions: July 15, 2025
• Notification of final decisions: September 15, 2025
• Tentative publication: Late 2025

Submission Information

Authors can submit their manuscripts via https://dl.acm.org/journal/tois. Submissions to this special issue will
follow the regular TOIS submission guidelines (https://dl.acm.org/journal/tois/author-guidelines). Submissions
must be accompanied by a cover letter containing all of the following: (1) Confirm that the paper is not currently
under submission at another journal or conference. (2) Confirm that the paper is substantially different from any
previously published work. (3) Confirm that none of the co-authors is a Guest Editor for this special session. (4)
Disclose possible conflicts of interest with Guests Editors. The review process will be single-blind. Strict policies will
be followed for plagiarism, submission confidentiality, reviewer anonymity, prior and concurrent paper submission
based on the guidelines.

Papers with a “Major Revision” decision should be resubmitted within three months, and with a “Minor Revision”
decision should be resubmitted within one month. Revised submissions must be accompanied with a detailed
response to reviewers explaining what revisions were implemented. The editors will conduct a second-round review
process and give the decision (accept or reject or need further revision) in one month.

For questions and further information, please contact Dr. Guglielmo Faggioli at guglielmo.faggioli@unipd.it.