Every field has a map.
We trace what is left uncharted.
Built for researchers who want to work where science needs them most, indexing 0 publications to expose critical, overlooked gaps.
Finding a gap means reading for months, and still guessing
A researcher choosing a direction has to hold an entire literature in their head, then bet that the absence they noticed is real and not simply a subject nobody needs.
- It does not scale. A literature review is bounded by how much one person can read, so the map is always a decade and a subfield wide at best.
- An absence is ambiguous. Nothing published on a topic can mean it is untouched, or that it is a non-problem. Reading alone cannot separate the two.
- The bias is invisible from inside. If your field studies English, the shape of what it skips never appears in the papers you read, only in the ones that were never written.
Millions of people who speak the language, divided by how much research exists on it.
English holds 3,577 papers in the same corpus.
Afieldrecordswhatitstudied.Itneverrecordswhatitskipped,sotheabsencehastobemeasuredinsteadofread.
Treat the blank space as a measurable object
If a field's coverage can be counted, then so can its absences. That turns choosing a research direction from an act of intuition into an argument with evidence attached.
An empty cell is not a missing record. It is a finding, and it should be scored, ranked and defended like any other result.
Nothing published is only interesting when comparable languages have solved the same task. That adjacency is what separates an opportunity from a non-problem.
A statistic with no path back to its records is a claim, not evidence. Each figure opens the papers that produced it.
Tags come from an explicit gazetteer, not model inference, so the same query returns the same answer and a reader can audit the reasoning.
The map, with its holes drawn in
Ten languages against eight common tasks. 6 of these 80 pairings have no indexed paper, and the hatched squares are exactly where a new contribution has no competition.
Four moves, from a phrase to a defensible shortlist
Each step is visible in the interface, and each one can be interrogated rather than taken on trust.
A pipeline you can audit end to end
Deterministic from source to score. No model sits in the chain, so the same question always returns the same answer.