by WxxShirley · Agent Tool · ★ 40
Agent-STAR This repository is the official implementation of our paper: Demystifying Reinforcement Learning for Long-Horizon Tool-Using Agents: A Comprehensive Recipe. We use TravelPlanner as a long-horizon tool-use testbed, where agents must iteratively call tools to satisfy multifaceted constraints, i.e., commonsense and hard constraints. We implement STAR [Data Synthesis → SFT → RL], a unified post-training pipeline that systematically studies the agentic RL design space across five axes: reward shaping, model scaling, data composition, algorithm selection, and environmental stability.
| Stars | 40 |
| Forks | 1 |
| Language | Python |
| Category | Agent Tool |
| Quality Score | 69.0210148342413/100 |
| Last Updated | 2026-08-23 |
| Created | 2026-03-21 |
| Platforms | python |
| Est. Tokens | ~15k |
These tools work well together with Agent-STAR for enhanced workflows:
Explore other popular agent tool tools:
Agent-STAR is [EMNLP 2026] Official implementation for paper "Demystifying Reinforcement Learning for Long-Horizon Tool-Using Agents: A Comprehensive Recipe". It is categorized as a Agent Tool with 40 GitHub stars.
Agent-STAR is primarily written in Python. It covers topics such as agent, agentic-rl, reinforcement-learning.
You can find installation instructions and usage details in the Agent-STAR GitHub repository at github.com/WxxShirley/Agent-STAR. The project has 40 stars and 1 forks, indicating an active community.