Wiki
Tags
About
FAQ
RSS
AIPOST

rejection sampling fine-tuning

Rejection sampling fine-tuning trains a model on selected successful outputs from a larger set of attempts.

1 article
Last mentioned

Rejection sampling fine-tuning generates multiple candidate outputs, retains those that meet a specified criterion, and uses them to fine-tune a model. Unlike reinforcement learning, which optimizes behavior using reward signals, it trains on the selected outputs as examples.

This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.

Articles covering this entry

Rejection sampling fine-tuning, or RFT, trains on successful attempts selected from a larger set of runs.

Read article

Related entries

Entries that often appear in the same articles.

Browse the wikiReport & corrections

prompt

A prompt is an instruction, question or contextual input that tells an AI model what task to perform.

77 articles

API

An API is a defined way for software programs to exchange data and use one another’s functions.

62 articles

Claude

Claude is Anthropic’s family of conversational AI models and services, first released in 2023.

62 articles
AIPOST
© 2026 AIPOST. All rights reserved.

AIPOST is an AI publication covering practical AI, AI security, performance, startups, health, ethics and industry news. No account is needed, and our privacy policy explains how we handle personal information.