Glossary · Phase 6 · Modern LLM engineering
Bradley-Terry model
A way to turn many "A beat B" judgments into one strength score per item, so you can predict how often any item beats any other.
Chess players get a rating from the games they win and lose. Nobody plays everyone, yet the ratings let you predict any matchup. You can do the same for answers, models or ad taglines with the Bradley-Terry model.
Try it now
A gap of 0 means a coin flip. A gap of 1 means about 73%. Comparisons turn into a single score for each answer.
Learn it properly
Instruction and preference data · The datasets that shape assistant behavior. About 30 minutes.
Other terms from the same lesson
- Cohen's kappa: How much two judges agree beyond what chance alone would give: 0 is no better than chance, 1 is perfect agreement.
- JSON Lines: A file format with one JSON record per line, the usual way to store training examples.
- position bias: Preferring whichever answer is shown first (or second) regardless of quality. Swapping the order and asking again exposes it.
- preference data: Pairs of answers to the same prompt where people (or a model) marked which one is better.
- sycophancy: A model telling people what they want to hear, such as agreeing with a wrong claim, because agreeable answers got rated higher in training.
Never trained an AI?
Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account.