Definição
Also called: pairwise comparison, side-by-side.
Comparing two or more model answers to the same prompt and saying which is better, often with a reason and a rating of how much better. These judgements are the main input to RLHF.
Used in: AI evaluator and rater roles.
Nas vagas abertas agora
116 of the open roles we list mention “preference ranking” or a word for it in the title or description, on 15 platforms (as of 3 November 2026). Of those that state an hourly rate (28 distinct postings), the median is $35/hr, and the middle half pay $20 to $50.
A maioria em: Prolific (34), Mercor (30), Turing (17), RWS TrainAI (11), Anyone AI (8), micro1 (3) e mais 9.
Digital Artistmicro1·United States
$21–65/hhá 1 dia
AI Data Specialist - French from CanadaRWS TrainAI·Canada·Horário flexível
$16.5/hhá 7 dias
AI Agent Tester - Real-World TasksJT AI Labs·15 países·Horário flexível
$25–45/hhá 10 dias
US Contract Attorney — LitigationMercor·United States·Horário flexível·Direito
$100–120/hhá 11 dias
Senior SWE - Python / TypeScriptWeekday·United States, Canada·Meio período
$100–150/hhá 15 dias
Quality Evaluator – Fluent English (Brazil)Anyone AI·Brazil·Período integral
$7–12/hhá 18 dias
Termos relacionados
EvaluationTesting what a model can do by having it answer a set of questions and scoring the answers, often by people working from a rubric.Checagem de fatosChecking each claim in a model's answer against reliable sources and marking what is wrong, unsupported or out of date.RLHFA way of training a model on people's judgements. ← Per approved hourTermo anterior, de A a Z. Project pause →Próximo termo, de A a Z. Glossário completoTodos os termos em uma página, de A a Z.Abrir →