agentic-grpo-longhorizon

by qiqihezh · Agent Tool · ★ 84

About agentic-grpo-longhorizon

Fixing GRPO training collapse in long-horizon multi-tool agents. A lightweight PRM-Lite + LATA joint approach achieves +37% over vanilla GRPO on τ-bench airline (50-task, multi-turn).

agentic-aigrpolong-horizonmulti-turn-agentsprocess-reward-modelqwenreinforcement-learningtau-benchtool-calling

Quick Facts

Stars84
Forks8
LanguagePython
CategoryAgent Tool
Quality Score52.2894523196696/100
Open Issues1
Last Updated2026-06-27
Created2026-04-29
Platformspython
Est. Tokens~13k

More Agent Tool Tools

Explore other popular agent tool tools:

View all Agent Tool tools →

Popular Python Agent Tools

Frequently Asked Questions

What is agentic-grpo-longhorizon?

agentic-grpo-longhorizon is Fixing GRPO training collapse in long-horizon multi-tool agents. A lightweight PRM-Lite + LATA joint approach achieves +37% over vanilla GRPO on τ-bench airline (50-task, multi-turn).. It is categorized as a Agent Tool with 84 GitHub stars.

What programming language is agentic-grpo-longhorizon written in?

agentic-grpo-longhorizon is primarily written in Python. It covers topics such as agentic-ai, grpo, long-horizon.

How do I install or use agentic-grpo-longhorizon?

You can find installation instructions and usage details in the agentic-grpo-longhorizon GitHub repository at github.com/qiqihezh/agentic-grpo-longhorizon. The project has 84 stars and 8 forks, indicating an active community.

View on GitHub → Browse Agent Tool tools