Dual-Scale World Models for LLM Agents Towards Hard-Exploration Problems
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914066683920384 |
|---|---|
| author | Kim, Minsoo Hwang, Seung-won |
| author_facet | Kim, Minsoo Hwang, Seung-won |
| contents | LLM-based agents have seen promising advances, yet they are still limited in "hard-exploration" tasks requiring learning new knowledge through exploration. We present GLoW, a novel approach leveraging dual-scale world models, maintaining a trajectory frontier of high-value discoveries at the global scale, while learning from local trial-and-error in exploration through a Multi-path Advantage Reflection mechanism which infers advantage-based progress signals to guide exploration. To evaluate our framework for hard-exploration, we tackle the Jericho benchmark suite of text-based games, where GLoW achieves a new state-of-theart performance for LLM-based approaches. Compared to state-of-the-art RLbased methods, our approach achieves comparable performance while requiring 100-800x fewer environment interactions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_24116 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Dual-Scale World Models for LLM Agents Towards Hard-Exploration Problems Kim, Minsoo Hwang, Seung-won Computation and Language LLM-based agents have seen promising advances, yet they are still limited in "hard-exploration" tasks requiring learning new knowledge through exploration. We present GLoW, a novel approach leveraging dual-scale world models, maintaining a trajectory frontier of high-value discoveries at the global scale, while learning from local trial-and-error in exploration through a Multi-path Advantage Reflection mechanism which infers advantage-based progress signals to guide exploration. To evaluate our framework for hard-exploration, we tackle the Jericho benchmark suite of text-based games, where GLoW achieves a new state-of-theart performance for LLM-based approaches. Compared to state-of-the-art RLbased methods, our approach achieves comparable performance while requiring 100-800x fewer environment interactions. |
| title | Dual-Scale World Models for LLM Agents Towards Hard-Exploration Problems |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.24116 |