From Word to World: Can Large Language Models be Implicit Text-based World Models?
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918370663727104 |
|---|---|
| author | Li, Yixia Wang, Hongru Qiu, Jiahao Yin, Zhenfei Zhang, Dongdong Qian, Cheng Li, Zeping Ma, Pony Chen, Guanhua Ji, Heng |
| author_facet | Li, Yixia Wang, Hongru Qiu, Jiahao Yin, Zhenfei Zhang, Dongdong Qian, Cheng Li, Zeping Ma, Pony Chen, Guanhua Ji, Heng |
| contents | Agentic reinforcement learning increasingly relies on experience-driven scaling, yet real-world environments remain non-adaptive, limited in coverage, and difficult to scale. World models offer a potential way to improve learning efficiency through simulated experience, but it remains unclear whether large language models can reliably serve this role and under what conditions they meaningfully benefit agents. We study these questions in text-based environments, which provide a controlled setting to reinterpret language modeling as next-state prediction under interaction. We introduce a three-level framework for evaluating LLM-based world models: (i) fidelity and consistency, (ii) scalability and robustness, and (iii) agent utility. Across five representative environments, we find that sufficiently trained world models maintain coherent latent state, scale predictably with data and model size, and improve agent performance via action verification, synthetic trajectory generation, and warm-starting reinforcement learning. Meanwhile, these gains depend critically on behavioral coverage and environment complexity, delineating clear boundry on when world modeling effectively supports agent learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_18832 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | From Word to World: Can Large Language Models be Implicit Text-based World Models? Li, Yixia Wang, Hongru Qiu, Jiahao Yin, Zhenfei Zhang, Dongdong Qian, Cheng Li, Zeping Ma, Pony Chen, Guanhua Ji, Heng Computation and Language Agentic reinforcement learning increasingly relies on experience-driven scaling, yet real-world environments remain non-adaptive, limited in coverage, and difficult to scale. World models offer a potential way to improve learning efficiency through simulated experience, but it remains unclear whether large language models can reliably serve this role and under what conditions they meaningfully benefit agents. We study these questions in text-based environments, which provide a controlled setting to reinterpret language modeling as next-state prediction under interaction. We introduce a three-level framework for evaluating LLM-based world models: (i) fidelity and consistency, (ii) scalability and robustness, and (iii) agent utility. Across five representative environments, we find that sufficiently trained world models maintain coherent latent state, scale predictably with data and model size, and improve agent performance via action verification, synthetic trajectory generation, and warm-starting reinforcement learning. Meanwhile, these gains depend critically on behavioral coverage and environment complexity, delineating clear boundry on when world modeling effectively supports agent learning. |
| title | From Word to World: Can Large Language Models be Implicit Text-based World Models? |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2512.18832 |