[CfP] ACM TOIS Efficiency in Neural IR

28 views
Skip to first unread message

Maria Maistro

unread,
Aug 12, 2022, 11:42:22 AM8/12/22
to Machine Learning News
Call for Papers - ACM Transactions on Information Systems
Special Section on Efficiency in Neural Information Retrieval


Full Call of Papers: https://dl.acm.org/journal/tois/calls-for-papers

Overview 🧐
--------------------------
The aim of this Special Section is to engage with researchers in Information Retrieval, Natural Language Processing, and related areas and gather insight into the core challenges in measuring, reporting, and optimizing all facets of efficiency in Neural Information Retrieval (NIR) systems, including time-, space-, resource-, sample- and energy- efficiency, among other factors.
This special section solicits perspectives from active researchers to advance our understanding of and to overcome efficiency challenges in NIR.
In particular, researchers are encouraged to examine the ever-growing model complexity through appropriate empirical analysis, to propose models that require less data, computational resources, and energy for training and fine-tuning with similarly efficient inference, to ask if there are meaningful simplifications of the existing training processes or model architectures that lead to comparable quality, and explore a multi-faceted evaluation of NIR models from quality to all dimensions of efficiency with standardized metrics.

Topics 🔍
--------------------------
We welcome submissions on the following topics, including but not limited to:
* Novel NIR models that reach competitive quality but are designed to provide efficient training or
inference;
* Efficient NIR models for decentralized IR tasks such as conversational search;
* Efficient NIR models for IR-related tasks such as question answering and recommender systems;
* Efficient NIR for resource-constrained devices;
* Scalability of NIR systems;
* Efficient NIR for text and cross-modal search;
* Strategies to optimize training or inference of existing NIR models;
* Sample-efficient training of NIR models;
* Efficiency-driven distillation, pruning, quantization, retraining, and transfer learning;
* Empirical investigation of the complexity of existing NIR models through an analysis of quality, interpretability, robustness, and environmental impact;
* Evaluation protocols for efficiency in NIR.

Important Dates 🔥
--------------------------
* Open for Submissions: Aug 1, 2022
* Submissions deadline: Dec 31, 2022
* First-round review decisions: Mar 31, 2023
* Deadline for minor revision submissions: Apr 30, 2023
* Deadline for major revision submissions: Jun 30, 2023
* Notification of final decisions: Jul 31, 2023
* Tentative publication: 2023

Guest Editors 📚
--------------------------
* Dr. Sebastian Bruch, Pinecone, United States of America
* Prof. Claudio Lucchese, Ca' Foscari University of Venice, Italy
* Dr. Maria Maistro, University of Copenhagen, Denmark
* Dr. Franco Maria Nardini, ISTI-CNR, Italy
Reply all
Reply to author
Forward
0 new messages