Research cases
Polished long-form cases. Demo cases are workflow examples, not excavation reports.
Research compilation from a published ICIT-style digitization — not a complete or licensed photographic corpus
RC-001
Does 033–705 form a recurring, directional compound in Mohenjo-daro inscriptions?
Illustrative workflow example. Requires verified corpus import for historical conclusions. In the demonstration set, 033 is followed by 705 more often than the reverse.
RC-002
Do particular signs show statistically significant terminal-position bias?
Terminal bias is measured as P(end)/P(any) with a minimum-occurrence threshold. Labels stay “ender-biased”, never “suffix” as fact.
RC-003
Do inscription templates differ more by object type than by seal motif?
A working functional hypothesis: tablets and seals may not encode the same kind of message. Motifs remain images.
RC-004
Can candidate numeral-like signs be linked to metrological or storage contexts?
Sign 342 is a numeral-like candidate. No calibrated link to the Indus weight series is claimed in the demo.
RC-005
Which sign variants may be allographs, directional mirrors or genuinely separate graphemes?
Merges must stay reversible. Visual similarity without positional similarity is not enough.