Can I read Enhancing Reasoning in LLMs with RLSP on EtoBox?

Enhancing Reasoning in LLMs with RLSP by Purva Natoo is a document available to read on EtoBox.

What is Enhancing Reasoning in LLMs with RLSP about?

This document discusses the transition of Large Language Models (LLMs) into Large Reasoning Models (LRMs) that can perform reasoning during inference. It introduces a post-training framework called Reinforcement Learning via Self-Play (RLSP), which enhances reasoning capabilities through a structured approach involving supervised fine-tuning, exploration rewards, and reinforcement learning. Empirical studies show that models trained with RLSP demonstrate significant improvements in reasoning abilities acros

Author

Purva Natoo

Language

EN