We invite you to participate in the SIGIR 2024 Workshop on Reaching Efficiency in Neural Information Retrieval (ReNeuIR). The workshop aims to foster discussion and collaboration on holistic evaluation of methods in the age of neural information retrieval (NIR), noting that effectiveness matters but so does the computational cost incurred to achieve it. Specific areas of interest include (but are not limited to):
-
Novel Neural IR (NIR) models that reach competitive quality but are designed to provide fast training or fast inference; -
Efficient NIR models for decentralized IR tasks such as conversational search; -
Strategies to speed up 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; and, -
Evaluation protocols for efficiency in NIR.
We accept both original submissions, as well as extensions or revisions to existing work. Papers can be submitted in two lengths – shorter poster submissions, and longer article submissions. Please see the website for more information: https://reneuir.org/cfp.html
-------------------------------------------------------------------------------
Important Dates
------------------------------------------------------------------------------- May 15, 2024: Paper submission deadline
June 7, 2024: Notification of acceptance for scientific papers
July 18, 2024: Workshop (co-located with SIGIR 2024 in Washington)
July 31, 2024: Final Proceedings Deadline (yes, after the conference)
-------------------------------------------------------------------------------
Questions? Please contact us! reneuir2024@easychair.org
Best regards,
ReNeuIR team