Publication
Ethics and eligibility
Most of this is the same set of rules the competition already imposes. The part people get wrong is the timing, not the paperwork.
Dates first
Required approvals have to be signed and dated before experimentation begins. A form signed afterward cannot be repaired later. If you have already started, say so now rather than at submission.
Human participants
Surveys count. Interviews count. Anything where a person supplies data about themselves counts, including your classmates and including anonymous responses. The approvals have to be in place before you collect anything, and no identifiable image or record of a participant appears in what we publish.
Vertebrate animals
Additional requirements apply and they are strict. Invertebrates sit outside the vertebrate rules but still require you to describe your handling honestly.
Hazardous materials, agents, and devices
Biological agents, controlled substances, and hazardous chemicals or equipment all pull in extra review. Describe what you used and under whose supervision.
Work done in an outside lab
Working in a university or company lab is allowed and is not something to hide. It does trigger additional requirements, and it makes the contributions statement more important rather than less: name what was yours and what was the lab's.
Disclosing AI use
Say where you used it and for what. Using a model to draft code, tidy prose, or explore ideas is normal and disclosable. Presenting generated text or generated results as your own work is not, and it is the kind of thing a reviewer asks about directly.
Plagiarism
Words and figures that are not yours are cited and attributed, every time. A figure redrawn from a paper still credits the paper.
Work presented or published elsewhere
Tell us before you submit. Some conferences hold rights that make republishing the full text a real problem rather than a hypothetical one. Where that is the case we publish a record that points at the authoritative version instead, which is better for you than a rights dispute later.
Asking
If you are not sure whether something applies to you, write to SciPath and ask before you start. That question has never once been a nuisance, and asking it late is how projects get disqualified.