Home/Data/Human Feedback

Human feedback

Preference ranking, RLHF, safety review, and cultural-alignment judgments — the same task type shown live on the home page.

A reviewer reads a prompt and one or more candidate responses and records a judgment: which response they'd accept, whether either is unsafe, whether the dialect used is authentic, whether a cultural reference lands correctly or misses. The judgment plus the reviewer's stated reason is the training signal.

Task types we run

  • Pairwise preference ranking (A vs. B, with rationale)
  • Absolute quality scoring against a rubric
  • Safety and harm review, including region-specific sensitivity that a global safety rubric misses
  • Dialect-authenticity checks — does this actually sound like natural Levantine Arabic, or does it read as MSA wearing a dialect's vocabulary
  • Hallucination review against a source document or known facts

How disagreement is handled

Reviewers disagree — that's expected, not an error state. Tasks with low inter-annotator agreement are routed to a senior reviewer rather than resolved by majority vote alone, and disagreement rates are reported back to you as part of the QA package, not hidden. See quality & methodology for how agreement is measured.