Call For Papers
We invite contributions that explore the intersection of ML, CV, and biological sciences.
The workshop aims to bring together researchers from diverse backgrounds to share insights, methodologies, and applications in Imageomics.
We welcome papers on a broad range of topics, including but not limited to:
- Foundation and Representation Learning for Biological Vision: build large-scale models that embed images and biological information (e.g., species names, trait ontologies) into a shared space.
- Data Standards, Benchmarks, and Open Resources: biological datasets that pose challenges to existing machine learning methods; trait-annotated image benchmarks, evaluation leaderboards, etc.
- Learning under Scarcity and Long-Tailed Distributions: few-/zero-shot recognition of rare or newly discovered species; learning for unbalanced fine-grained classification.
- Multi-sensor Fusion: Combine the representations of images, sounds, DNA, field notes, etc.
- Temporal and Behavioral Reasoning from Video: recognition of behavioral states in long video sequences.
- Hierarchy-Aware Learning: embeddings and objectives that respect hierarchical information, such as evolutionary trees or taxonomic labels.
- Trustworthy, Explainable, and Uncertainty-Aware Bio-Vision: saliency maps tied to morphological and functional concepts; calibrated confidence for decision-making.
- Human-in-the-Loop and Citizen-Science Annotation: active-learning pipelines for experts and volunteers, balancing annotation cost and data quality.
We also encourage submissions related to broader biological topics, such as
biomedical, evolution, conservation, taxonomy, and biodiversity.
Submission Guidelines
We invite submissions including:
- Short papers presenting ongoing research or work submitted to other venues (up to 5 pages, excluding references)
- Full papers that have been published at a peer-reviewed venue after January 1st, 2024 (up to 9 pages, excluding references)
All the submissions must be anonymous and follow the NeurIPS style.
An appendix (without page limits) is allowed after references and should be submitted in the same file as the main text.
The full paper submissions must not violate any dual-submission or anonymity policies of the accepted venues.
The accepted papers will NOT be archived in NeurIPS proceedings. This allows the authors to extend their work afterward and submit it to other venues.