logo
How it worksCoursesResearch CommunitiesBenefitsAbout Us
Schedule Demo
Learn Before
  • Classification of Reward Models for LLM Reasoning

Concept icon
Concept

Outcome Reward Models

An outcome reward model is a type of verifier used in reinforcement learning for LLMs that evaluates the final answer of a reasoning process. It assesses the correctness or overall quality of the end result, providing a reward signal based solely on this final evaluation.

0

1

Concept icon
Updated 2026-05-06

Contributors are:

Gemini AI
Gemini AI
🏆 3

Who are from:

Google
Google
🏆 3

References


  • Reference of Foundations of Large Language Models Course

  • Reference of Foundations of Large Language Models Course

Tags

Ch.5 Inference - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences

Related
  • Outcome Reward Models

    Concept icon
  • Process Reward Models

    Concept icon
  • Rule-Based Reward Models for Reasoning

    Concept icon
  • A team is training a language model to solve multi-step logic puzzles. Their training system automatically reviews each line of the model's generated reasoning. If a line represents a valid deductive step, it receives a positive score. If a line contains a logical fallacy or contradicts a previous statement, it receives a negative score, and the evaluation stops. The total score for the entire reasoning path is then used to update the model. Which classification best describes this type of feedback mechanism?

  • Selecting a Reward Model for a Math Tutoring LLM

  • Match each description of a feedback mechanism for training a reasoning model with the most appropriate classification.

Learn After
  • A team is training a language model to act as a programming assistant that writes code to solve specific problems. Their training method involves running the code generated by the model. If the code executes without errors and produces the correct output for a set of predefined tests, the model receives a high reward. If the code fails to execute or produces the wrong output, it receives a low reward. The system does not evaluate the elegance, efficiency, or style of the code itself, only the final result of its execution. Which of the following statements best characterizes this evaluation approach?

  • Analyzing a Reward System's Weakness

  • Evaluating a Reward System for an AI Tutor

logo 1cademy1Cademy

Optimize Scalable Learning and Teaching

How it worksCoursesResearch CommunitiesBenefitsAbout Us
TermsPrivacyCookieGDPR

Contact Us

iman@honor.education

Follow Us




© 1Cademy 2026

We're committed to OpenSource on

Github