GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918260369260544 |
|---|---|
| author | Guo, Jiacheng Yang, Ling Chen, Peter Xiao, Qixin Wang, Yinjie Juan, Xinzhe Qiu, Jiahao Shen, Ke Wang, Mengdi |
| author_facet | Guo, Jiacheng Yang, Ling Chen, Peter Xiao, Qixin Wang, Yinjie Juan, Xinzhe Qiu, Jiahao Shen, Ke Wang, Mengdi |
| contents | Training capable Large Language Model (LLM) agents is critically bottlenecked by the high cost and static nature of real-world interaction data. We address this by introducing GenEnv, a framework that establishes a difficulty-aligned co-evolutionary game between an agent and a scalable, generative environment simulator. Unlike traditional methods that evolve models on static datasets, GenEnv instantiates a dataevolving: the simulator acts as a dynamic curriculum policy, continuously generating tasks specifically tailored to the agent's ``zone of proximal development''. This process is guided by a simple but effective $α$-Curriculum Reward, which aligns task difficulty with the agent's current capabilities. We evaluate GenEnv on five benchmarks, including API-Bank, ALFWorld, BFCL, Bamboogle, and TravelPlanner. Across these tasks, GenEnv improves agent performance by up to \textbf{+40.3\%} over 7B baselines and matches or exceeds the average performance of larger models. Compared to Gemini 2.5 Pro-based offline data augmentation, GenEnv achieves better performance while using 3.3$\times$ less data. By shifting from static supervision to adaptive simulation, GenEnv provides a data-efficient pathway for scaling agent capabilities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_19682 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators Guo, Jiacheng Yang, Ling Chen, Peter Xiao, Qixin Wang, Yinjie Juan, Xinzhe Qiu, Jiahao Shen, Ke Wang, Mengdi Computation and Language Training capable Large Language Model (LLM) agents is critically bottlenecked by the high cost and static nature of real-world interaction data. We address this by introducing GenEnv, a framework that establishes a difficulty-aligned co-evolutionary game between an agent and a scalable, generative environment simulator. Unlike traditional methods that evolve models on static datasets, GenEnv instantiates a dataevolving: the simulator acts as a dynamic curriculum policy, continuously generating tasks specifically tailored to the agent's ``zone of proximal development''. This process is guided by a simple but effective $α$-Curriculum Reward, which aligns task difficulty with the agent's current capabilities. We evaluate GenEnv on five benchmarks, including API-Bank, ALFWorld, BFCL, Bamboogle, and TravelPlanner. Across these tasks, GenEnv improves agent performance by up to \textbf{+40.3\%} over 7B baselines and matches or exceeds the average performance of larger models. Compared to Gemini 2.5 Pro-based offline data augmentation, GenEnv achieves better performance while using 3.3$\times$ less data. By shifting from static supervision to adaptive simulation, GenEnv provides a data-efficient pathway for scaling agent capabilities. |
| title | GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2512.19682 |