Group-Aware Reinforcement Learning for Output Diversity in Large Language Models

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Anschel, Oron, Shoshan, Alon, Botach, Adam, Hakimi, Shunit Haviv, Gendler, Asaf, Baruch, Emanuel Ben, Bhonker, Nadav, Kviatkovsky, Igor, Aggarwal, Manoj, Medioni, Gerard
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914160249405440
author Anschel, Oron
Shoshan, Alon
Botach, Adam
Hakimi, Shunit Haviv
Gendler, Asaf
Baruch, Emanuel Ben
Bhonker, Nadav
Kviatkovsky, Igor
Aggarwal, Manoj
Medioni, Gerard
author_facet Anschel, Oron
Shoshan, Alon
Botach, Adam
Hakimi, Shunit Haviv
Gendler, Asaf
Baruch, Emanuel Ben
Bhonker, Nadav
Kviatkovsky, Igor
Aggarwal, Manoj
Medioni, Gerard
contents Large Language Models (LLMs) often suffer from mode collapse, repeatedly generating the same few completions even when many valid answers exist, limiting their diversity across a wide range of tasks. We introduce Group-Aware Policy Optimization (GAPO), a simple extension of the recent and popular Group Relative Policy Optimization (GRPO) that computes rewards over the group as a whole. GAPO enables learning from the group-level properties such as diversity and coverage. We demonstrate GAPO using a frequency-aware reward function that encourages uniform sampling over valid LLM completions, and show that GAPO-trained models produce valid and more diverse model responses. Beyond this setup, GAPO generalizes to open-ended prompts and improves response diversity without compromising accuracy on standard LLM benchmarks (GSM8K, MATH, HumanEval, MMLU-Pro). Our code will be made publicly available.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12596
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Group-Aware Reinforcement Learning for Output Diversity in Large Language Models
Anschel, Oron
Shoshan, Alon
Botach, Adam
Hakimi, Shunit Haviv
Gendler, Asaf
Baruch, Emanuel Ben
Bhonker, Nadav
Kviatkovsky, Igor
Aggarwal, Manoj
Medioni, Gerard
Computation and Language
Artificial Intelligence
Machine Learning
Large Language Models (LLMs) often suffer from mode collapse, repeatedly generating the same few completions even when many valid answers exist, limiting their diversity across a wide range of tasks. We introduce Group-Aware Policy Optimization (GAPO), a simple extension of the recent and popular Group Relative Policy Optimization (GRPO) that computes rewards over the group as a whole. GAPO enables learning from the group-level properties such as diversity and coverage. We demonstrate GAPO using a frequency-aware reward function that encourages uniform sampling over valid LLM completions, and show that GAPO-trained models produce valid and more diverse model responses. Beyond this setup, GAPO generalizes to open-ended prompts and improves response diversity without compromising accuracy on standard LLM benchmarks (GSM8K, MATH, HumanEval, MMLU-Pro). Our code will be made publicly available.
title Group-Aware Reinforcement Learning for Output Diversity in Large Language Models
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2511.12596