Foundation Models for Science Big Data
Time: Dec 15-18, 2024
Location: Washington D.C., USA
Introduction
The capabilities of foundation models for science big data span a wide spectrum, from the atomic level, where it solves partial differential equations for quantum systems, to the molecular level, predicting chemical or protein structures, and further extending to societal predictions like infectious disease outbreaks. However, there remain significant challenges to unlock the full potential of foundation models for scientific discovery, such as ensuring trustworthiness, achieving personalization, and adapting to multi-modal data representation. This workshop welcomes a wide range of contributions involving developing, analyzing, or applying foundation models for science big data spanning across various domains (e.g., biomedicine, chemistry, math, physics, material science, and geography) and various modalities (e.g., text, images, tables, graphs, and time series).
Submission
We accept both long papers (up to 10 pages in IEEE 2-column format, including references) and short papers (up to 5 pages in IEEE 2-column format, including references). Papers should be formatted to IEEE Computer Society Proceedings Manuscript Formatting Guidelines. Submissions that do not adhere to the format requirements will be rejected without review.
The submission system: On-Line Paper Submission.
Review
The review process is single-blind, meaning that reviewers remain anonymous, but authors are not. All papers accepted for this workshop will be included in the Workshop Proceedings published by the IEEE Computer Society Press and made available at the IEEE Big Data 2024 Conference. Each accepted paper is required to have at least one author register and present the work at the workshop during the IEEE Big Data 2024 Conference.
Authorship
The list of authors for submissions must comprise individuals who have made significant contributions to the work presented. The author list, including the order, cannot be altered after submission.
Ethics Statement
We strongly encourage all authors to include an explicit ethics statement regarding the broader impact of their work or any other ethical considerations. This statement should be positioned after the conclusion but before the references. The ethics statement will not be counted toward the page limits for submissions.
Supplementary Materials
You have the option to include unlimited pages of appendices at the end of your submission, following the references. Appendices should be used to provide additional support for the findings and points presented in the main content but should not be necessary to grasp the core contributions of your work.
For software (source codes) or data, as our submission process is single-blind, we encourage you to host them in public repositories, such as GitHub, and provide links for reference.
Research Topics
We consider a broad range of subject areas focused on foundation models for science big data. A non-exhaustive list of topics of interest includes:
- Science Domains: biomedicine, chemistry, math, physics, material science, geography, …
- Data Modalities: text, images, tables, graphs, time series, domain-specific data, …
- Model Life-cycle: development, analysis, and applications, …
- Evaluation: benchmarks, scalable oversight, evaluation protocols and metrics, human and/or machine evaluation, …
- Trustworthiness: factuality, groundedness, uncertainty, explainability, bias and safety, security, privacy, robustness, …
- Compute Efficiency: distillation, compression, quantization, sample efficient methods, memory efficient methods, …
- Engineering: distributed training and inference on different hardware setups, training dynamics, optimization instability, …
- Learning Algorithms: learning, unlearning, meta-learning, model mixing methods, continual learning, …
- Inference Algorithms: decoding algorithms, reasoning algorithms, search algorithms, planning algorithms, …
- Tools and Code: pre-trained foundation models, integration with tools and APIs, LM-driven software engineering …
Important Dates
- Oct 1, 2024: Due date for full workshop papers submission
- Nov 4, 2024: Notification of paper acceptance to authors
- Nov 20, 2024: Camera-ready of accepted papers
- Dec 15-18, 2024: Workshop date (TBD)