Call for papers · Open

Call for Papers

SIMBIOCHEM is currently accepting submissions — non-archival short papers (5–8 pages) and abstracts on machine learning for simulation in biology and chemistry. Read this page top to bottom; the anonymisation and COI rules matter.

Submit on OpenReview

Deadline 29 August 2026, 11:59 PM UTC

Submissions open

Now on OpenReview

Submission deadline

August 29, 2026

11:59 PM UTC (UTC-0)

Author notification

September 29, 2026

AoE

Workshop day

Dec 11 or 12, 2026

Exact day TBC by NeurIPS

Camera-ready (accepted papers only): after notification — the exact date is announced with decisions. All deadlines are on OpenReview.

About the workshop

What the workshop is about

Machine learning has transformed biology and chemistry, yet many models miss essential dynamics. First-principles methods like molecular dynamics offer physical grounding but remain costly. SIMBIOCHEM bridges that divide — integrating scalable ML with rigorous physical simulation to make methods faster, and more accurate, reliable and grounded in science.

How can simulation, generative models, and agentic AI be fused into physics-aligned systems that learn from reality and accelerate biological and chemical discovery?
The guiding question

Who can submit

A broad range of contributions across ML, computational chemistry, biophysics and materials science. We especially welcome work on how ML can accelerate physical simulation, and how simulation can inform and strengthen ML — from datasets and benchmarks to new models and algorithms.

Topics of interest

  • • Physically-grounded architectures (symmetries, conservation laws).
  • • Differentiable simulation & inverse design.
  • • Next-generation learned potentials.
  • • Data efficiency & scalability, foundation models, agentic ML.
  • • Standardisation & reproducibility: benchmarks, datasets, metrics.
1

What to submit

Submission is entirely electronic, via the SIMBIOCHEM venue on OpenReview.

What we welcome

SIMBIOCHEM is a place for scientific discussion, not only for finished results. We actively welcome early-stage work: methods that look promising but are not yet validated in the lab, partial or negative results others can learn from, and early ideas that could steer where the community goes next.

Experienced and early-career researchers are equally welcome, from academia and from industry. Work being early is not a mark against it here — we only ask that you are clear about what has been shown and what has not.

Academic rigour still applies. Every submission is reviewed on four things:

Novelty
A new idea, or a genuinely new angle on a known one.
Impact
If it holds up, does it matter to the field?
Correctness
Are the claims supported by what is actually shown?
Clarity
Can a reader follow the method and the reasoning?

Papers (5–8 pages)

  • • 5–8 pages, excluding references, appendices and data-availability statements (which may take unlimited space).
  • • Reviewers reserve the right to stop reading beyond 8 pages; extra content is at the author’s risk.
  • • Six spotlight papers are selected from accepted submissions.
  • • Must be fully anonymised using the provided template (see step 2).

Camera-ready (if accepted)

Accepted papers will be invited to submit a camera-ready version. Although the workshop is non-archival, camera-ready submissions are collated into a proceedings page on this website for download — with an opt-out at submission.

Posters (if accepted)

  • • All accepted papers may present a poster in person.
  • 24″ (W) × 36″ (H), portrait, on lightweight paper (NeurIPS workshop size).
  • • Mount with provided command strips only (command strips and painters tape are provided in the room). No adhesive-backed self-sticking posters or anything that could damage walls or boards.
  • • Please take your poster down after your session.
  • • NeurIPS will suggest preferred poster-printing options nearer the conference.

Slides (if spotlighted)

Six spotlight papers are chosen by the organisers with the committees. Each gets a 10-minute talk; presenters submit slides at least two days before the workshop.

Journal fast-track & awards

The organising committee is reaching out to journals about fast-track publication opportunities and a best-paper award for top accepted submissions. Any such pathway remains at the journal’s discretion and independent of the workshop review.

Non-archival. There are no proceedings and no rebuttal phase. We welcome work previously presented at other workshops or conferences, or currently under review elsewhere.

2

Get the template & prepare an anonymised PDF

Use the official NeurIPS 2026 LaTeX style and submit anonymised — omit the final and preprint options. For Overleaf, download the zip and upload it as a new project.

Anonymise carefully — desk-reject otherwise

  • • Submissions must be fully anonymised as per the template instructions. Do not include author names, affiliations or acknowledgements.
  • • Remove any identifying links — GitHub repositories, personal websites, and images/figures (or their metadata) that could reveal your identity.
  • Failing to submit an anonymised paper will cause desk rejection. Acknowledgements may be added only in the camera-ready version after acceptance.
  • • Because the workshop is non-archival, reviewers are explicitly instructed to evaluate only the submitted PDF and not to search for external sources — please help preserve double-blind review by removing anything that could deanonymise you.

Research integrity — no duplicate submissions

Authors must comply with research-integrity guidelines. Submitting the same paper more than once — tweaked duplicates, or multiple versions of the same work — is misconduct. We run mechanisms to detect this, and any such attempt (or any other attempt to outsmart the review process) will be reported to the organisers, OpenReview and NeurIPS as misconduct, and will be desk-rejected.

SIMBIOCHEM · NeurIPS 2026 template

Includes neurips_2026.tex, neurips_2026.sty and the required checklist.tex.

Checklist before you submit

  • • 5–8 pages (references & appendices excluded).
  • • Fully anonymised, no identifying links or images.
  • • US-Letter page size; embed Type 1 / TrueType fonts.
  • • COI declaration completed on OpenReview (step 3).
  • • The paper checklist is encouraged but not enforced — please complete the AI/LLM disclosure item.
3

Declare your conflicts of interest

Required this year — taken seriously

NeurIPS 2026 distinguishes domain conflicts, which come from the past three years of Education & Career History, from personal conflictswith individuals. Keep every author’s OpenReview profile current so these declarations can support conflict-safe review assignment.

Submission eligibility

Under the NeurIPS 2026 workshop guidance, organisers and anyone with a personal conflict of interest with an organiser may not submit. NeurIPS gives an organiser’s PhD student or postdoc as examples. A shared current or recent institutional affiliation is a domain conflict; it does not by itself make an author ineligible, but the conflicted organiser will not assess that submission.

How NeurIPS 2026 defines the two types of conflict

  • Domain conflict. Affiliations in the past three years, entered in OpenReview Education & Career History. Include all affiliations, such as consulting and sabbaticals.
  • Family or close personal relationship. This is a personal conflict.
  • PhD advisor / advisee. A PhD advisor–advisee relationship in either direction is a personal conflict.
  • Recent original-research co-authorship. Co-authorship on an original research article within the past three years is a personal conflict. Perspective pieces do not count.
  • Other exceptional conflicts. If another relationship would significantly compromise review fairness, declare it through OpenReview or contact the workshop organisers for guidance.

Organisers covered by the submission restriction

The restriction on submitting because of a personal conflict applies to the organisers listed below, not to members of the advisory committee.

Bruno TrentiniUniversity of Oxford · NVIDIA
Emine KucukbenliNVIDIA
Jigyasa NigamMIT
Max SecorNovo Nordisk
Ole WintherU. of Copenhagen · DTU · Raffle.ai
Runzhong WangMIT (Coley Lab)

Review, prizes & reviewing

How review works

Review process

Up to three double-blind reviews per submission, with reviewer loads capped at three. Approve/reject recommendations are required; detailed comments are encouraged. There is no rebuttal phase.

Recognition & prizes

Best papers are selected by the organisers and advisors on novelty, impact and presentation. The committee is reaching out to sponsors and journals about additional prizes and fast-track opportunities.

Reviewers wanted

We welcome reviewers of all levels of seniority; reviewers are listed on the website.

Join the Programme Committee

Use of AI in reviewing

Reviewers may use AI to assist their reviews in accordance with NeurIPS guidelines, under these guiding principles:

  • Human judgement is augmented, not replaced. The tool assists reviewers and does not replace any human in the review process.
  • Informed author consent. A paper is included in any AI-assistance experiment only if its authors opt in.
  • Confidentiality & privacy are upheld to the highest standards.
  • Disclosure is encouraged. If you used AI to help write a review, we would rather you said so.

Reviewers are expected to be fair, to stick to the paper under review, and not to search for information that could identify authors. We encourage using AI wisely per NeurIPS guidelines. Papers that are clearly AI-generated and of poor quality may be desk-rejected. If authors believe a reviewer used AI unwisely, they may report the review to the organisers.

4

Submit on OpenReview

Once you have prepared an anonymised PDF with the template and completed the COI declaration, submit through the SIMBIOCHEM OpenReview portal before the deadline.

Anonymised PDF, no identifying links NeurIPS 2026 template COI declaration complete
Submit on OpenReview

Deadline 29 August 2026, 11:59 PM UTC · questions to workshop@simbiochem.com