Language-Guided World Models: A Model-Based Approach to AI Control
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910589547184128 |
|---|---|
| author | Zhang, Alex Nguyen, Khanh Tuyls, Jens Lin, Albert Narasimhan, Karthik |
| author_facet | Zhang, Alex Nguyen, Khanh Tuyls, Jens Lin, Albert Narasimhan, Karthik |
| contents | This paper introduces the concept of Language-Guided World Models (LWMs) -- probabilistic models that can simulate environments by reading texts. Agents equipped with these models provide humans with more extensive and efficient control, allowing them to simultaneously alter agent behaviors in multiple tasks via natural verbal communication. In this work, we take initial steps in developing robust LWMs that can generalize to compositionally novel language descriptions. We design a challenging world modeling benchmark based on the game of MESSENGER (Hanjie et al., 2021), featuring evaluation settings that require varying degrees of compositional generalization. Our experiments reveal the lack of generalizability of the state-of-the-art Transformer model, as it offers marginal improvements in simulation quality over a no-text baseline. We devise a more robust model by fusing the Transformer with the EMMA attention mechanism (Hanjie et al., 2021). Our model substantially outperforms the Transformer and approaches the performance of a model with an oracle semantic parsing and grounding capability. To demonstrate the practicality of this model in improving AI safety and transparency, we simulate a scenario in which the model enables an agent to present plans to a human before execution, and to revise plans based on their language feedback. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_01695 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Language-Guided World Models: A Model-Based Approach to AI Control Zhang, Alex Nguyen, Khanh Tuyls, Jens Lin, Albert Narasimhan, Karthik Computation and Language Artificial Intelligence Machine Learning This paper introduces the concept of Language-Guided World Models (LWMs) -- probabilistic models that can simulate environments by reading texts. Agents equipped with these models provide humans with more extensive and efficient control, allowing them to simultaneously alter agent behaviors in multiple tasks via natural verbal communication. In this work, we take initial steps in developing robust LWMs that can generalize to compositionally novel language descriptions. We design a challenging world modeling benchmark based on the game of MESSENGER (Hanjie et al., 2021), featuring evaluation settings that require varying degrees of compositional generalization. Our experiments reveal the lack of generalizability of the state-of-the-art Transformer model, as it offers marginal improvements in simulation quality over a no-text baseline. We devise a more robust model by fusing the Transformer with the EMMA attention mechanism (Hanjie et al., 2021). Our model substantially outperforms the Transformer and approaches the performance of a model with an oracle semantic parsing and grounding capability. To demonstrate the practicality of this model in improving AI safety and transparency, we simulate a scenario in which the model enables an agent to present plans to a human before execution, and to revise plans based on their language feedback. |
| title | Language-Guided World Models: A Model-Based Approach to AI Control |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2402.01695 |