Two words doing very different work
The criterion at 8 CFR 204.5(h)(3)(v) concerns original contributions of major significance in the field. Originality is the low bar: peer review already tested it, and a published paper is by definition a contribution that was not there before.
"Major significance" is where petitions succeed or fail. It is a claim about impact beyond your own work and your own institution — that the field changed in some traceable way because of what you did.
This is why a strong publication list is not automatically a strong showing here. Papers that were published, cited a handful of times by collaborators, and then left alone are original. Demonstrating that they were significant requires evidence of a different kind.
Originality is established by publication. Significance has to be shown by what others did next.
What evidence actually demonstrates significance
Independent citation is the most direct evidence — not raw counts, but citations from researchers with no connection to you, ideally showing that your result was built on rather than merely listed. A paper cited fifteen times by fifteen unconnected groups who each extended the work says more than one cited eighty times by your own lab.
Adoption is stronger still. A method other groups now use, a dataset others analyse, a tool that appears in other people's pipelines, a result that changed practice or entered a guideline — these show significance in a way citation counts only approximate.
Independent expert letters matter here more than anywhere else in the petition, and their value is inversely proportional to how generic they are. A letter from someone who has never worked with you, explaining specifically how your finding changed what their group does, is worth many pages of general praise from a former supervisor.
Field context is part of the argument
Citation norms vary enormously between disciplines. A count that is exceptional in pure mathematics would be ordinary in immunology, and an officer is not obliged to know that. Making the comparison explicit — against field baselines, against percentile data for the subject category — turns a bare number into an argument.
The same applies to venue. Explaining what a journal's standing is within your specific subfield, with evidence rather than adjectives, does work that the journal name alone will not do.
None of this can be manufactured after the fact, which is the honest and uncomfortable part. Significance accrues over years. What can be done deliberately is choosing problems whose solutions other people will need, and publishing where those people will find them.