Generative AI in Genomics (Gen2): Barriers and Frontiers

ICLR 2026 Workshop  |  April 26-27  |  Rio de Janeiro, Brazil

Contact: genai-in-genomics@googlegroups.com


Call for Papers


Gen2 is designed to sharply focus on domain-specific roadblocks and open challenges that must be addressed to accelerate the impact of GenAI in genomics. We welcome submissions of unpublished work and works-in-progress on a range of topics at this intersection, as listed in the "Topics of Interest" section below.

We invite submissions through our OpenReview portal in two tracks:

  • Full Workshop Papers (4-8 pages), and
  • Tiny Papers (2-4 pages),
latter of which is designed to encourage the communication of early findings and lower the barrier to entry for early-career researchers.

Both tracks are considered non-archival to avoid conflict with future publication. The review process is double-blind. We ask that authors strictly follow the LaTeX templates provided below and submit anonymized papers.
Please review the submission instructions on this page and reach out to genai-in-genomics@googlegroups.com if you have any questions.

Reciprocal Review: To ensure a fair and high-quality review process, we require that each submission nominate at least one author to review up to 3 submissions between February 5-25 via the submission page.

Key Dates


  • Submissions open: Sun, 28 Dec 2026
  • Submissions close: Thu, 5 Feb 2026, 11:59PM UTC-12 (Anywhere on Earth)
  • Notification: Sun, 1 March 2026, 11:59PM UTC-12 (Anywhere on Earth)
  • Camera-ready deadline: Tue, 10 March 2026, 11:59PM UTC-12 (Anywhere on Earth)
  • Workshop: Sun or Mon, 26 or 27 Apr 2026 (TBD)

Topics of Interest


We welcome submissions covering position papers, new datasets and tasks, novel GenAI method development, and applications of existing models in genomics. Topics of interest include, but are not limited to:

Core Genomics Applications

Single-cell -omics:
  • Multi-omic integration,
  • conditional and cross-modal generation,
  • single cell trajectory simulation,
  • perturbation effect modeling,
  • target identification,
  • virtual cell models.
Spatial transcriptomics and tissue modeling:
  • Generative models for in silico tissue reconstruction,
  • cell-cell interaction modeling,
  • self-supervised learning of tissue domains,
  • spatial perturbation response.
Regulatory genomics and epigenomics:
  • Sequence-to-function deep learning models of genome regulations,
  • DNA language models with multi-modality,
  • design of regulatory elements for cell-type-specific activity,
  • DNA and RNA sequence generation.
Population genomics:
  • Ancestry-aware generative models,
  • rare variant simulation,
  • generative modeling for variant effect prediction,
  • genotype-phenotype modeling.

Methodological Themes

Adapting generative models to biological data:
  • Diffusion- and flow-based generative models,
  • transformers,
  • VAEs and GANs etc tailored for biological data and tasks.
Pretrained foundation models for genomics:
  • Large-scale self-supervised learning across multiple scales and data modalities,
  • genome-scale masked modeling.
Latent space interpretation:
  • Interpretation of large-scale foundation models,
  • enforcing disentanglement of biological factors,
  • linking latent variables to known biology.
Data-efficient generation:
  • Techniques to generate from few-shot data or under noisy, incomplete annotations common in genomics.
Evaluation and benchmarks:
  • Designing suitable benchmarking tasks and evaluation criteria,
  • re-evaluating existing models with robust benchmarks and metrics.

Submission Instructions


All submissions will be handled on openreview.net:

Submit on OpenReview


We offer two submission tracks :

Regular Papers

Regular paper submissions may be up to 8 pages, excluding Software and Data, References, and Appendix.
Submissions that violate page limits or anonymization policies will be desk rejected.

Please use the Gen2 Workshop LaTeX Template when preparing your submission or make copy of this Overleaf project

.

Tiny Papers

Tiny paper submissions may be up to 4 pages, excluding Software and Data, References, and Appendix.
Submissions that violate page limits or anonymization policies will be desk rejected.

Please use the Gen2 Workshop LaTeX Template when preparing your submission or make copy of this Overleaf project . We ask that you use gen2_iclr2026tiny.sty file for the Tiny Papers submissions, but we will not penalize submissions that use the default style file.

In line with ICLR regulations, for this track we especially encourage submissions that present:

  • An implementation and evaluation of an unpublished but simple idea
  • A modest yet self-contained theoretical result
  • A follow-up experiment or re-analysis of a previously published paper
  • A fresh perspective on an existing publication

Since 2025, ICLR has discontinued the separate “Tiny Papers” track, and is instead requiring each workshop to accept short (3–5 pages in ICLR format, exact page length to be determined by each workshop) paper submissions, with an eye towards inclusion; see ​​https://iclr.cc/Conferences/2025/CallForTinyPapers for a history of the ICLR tiny papers initiative.
Authors of these papers will be earmarked for potential funding from ICLR, but need to submit a separate application for Financial Assistance that evaluates their eligibility. This application for Financial Assistance to attend ICLR 2026 will become available on https://iclr.cc/Conferences/2026/ at the beginning of February and close early March.

Originality of Submission


Submitted work must be original and not previously published. Submissions that have appeared on preprint servers, such as arXiv, or other non-archival workshops are permitted.

Review Process


The review process is double-blind. Please ensure your submissions are anonymized.

For both tracks, authors may include an unlimited number of appendix pages after the bibliography. However, please keep in mind that the reviewers are not required to read the appendix. Therefore, please try to communicate your core findings and their importance in the main text.

Use of Large Language Models (LLMs)


In line with ICLR regulations, the use of large language models is permitted as a general-purpose assistance tool. However, if an LLM plays a significant role in research ideation and/or writing, such that it could reasonably be considered a contributor, authors must disclose its use. The specific role of the LLM should be described in a dedicated LLM Usage section, which may be included in the appendix and does not count toward the page limit. Failure to disclose significant LLM usage may result in desk rejection.

For the Tiny Paper track, in line with ICLR regulations, fully AI-generated papers are not permitted in this submission format. AI assistance is allowed, but submissions must be primarily human-authored and reflect original thought and analysis. Authors are encouraged to consult guidance from NYU and MIT regarding the distinction between AI-generated and AI-assisted work.

Presentation and Publications


All accepted papers will be presented as posters during the workshop. A subset of papers from the Regular Paper track will be selected for spotlight oral presentations based on reviewer feedback.

While the workshop is non-archival, authors of accepted papers will have the option to make their paper, poster, or presentation slides available through the workshop website and the Gen2 OpenReview venue.

We will encourage presenters of accepted submissions to attend the workshop in person, while accommodating virtual attendance in cases of visa issues or other exceptional circumstances. In such cases, we will coordinate in advance to collect pre-recorded presentations to broadcast during the event.

Frequently Asked Questions (FAQ)


What does "non-archival" mean?

This workshop is non-archival, meaning that accepted papers will not appear in official proceedings and will not preclude future submission of the same or extended work to archival conference or journal venues.

Can I submit the same work to another workshop?

Yes, concurrent submission of the same or related work to other venues is permitted. However, please double-check the policies of any other venues to which you are submitting.

Can I submit a work at the workshop that has already been accepted elsewhere?

As this workshop is non-archival, submissions of work that have been accepted but not yet formally published are permitted.