Concept

Training-Based Methods for Scaling LLM Reasoning

Training-based methods scale Large Language Model reasoning by further training or fine-tuning the model parameters to explicitly improve its reasoning abilities. For instance, a model might undergo supervised fine-tuning on datasets containing reasoning examples, such as math problems accompanied by step-by-step solutions.

0

1

Updated 2026-05-06

Contributors are:

Who are from:

Tags

Ch.5 Inference - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences