Unifying Deep Predicate Invention with Pre-trained Foundation 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_ | 1866917159555301376 |
|---|---|
| author | Wang, Qianwei Li, Bowen Luo, Zhanpeng Xu, Yifan Gray, Alexander Silver, Tom Scherer, Sebastian Sycara, Katia Xie, Yaqi |
| author_facet | Wang, Qianwei Li, Bowen Luo, Zhanpeng Xu, Yifan Gray, Alexander Silver, Tom Scherer, Sebastian Sycara, Katia Xie, Yaqi |
| contents | Long-horizon robotic tasks are hard due to continuous state-action spaces and sparse feedback. Symbolic world models help by decomposing tasks into discrete predicates that capture object properties and relations. Existing methods learn predicates either top-down, by prompting foundation models without data grounding, or bottom-up, from demonstrations without high-level priors. We introduce UniPred, a bilevel learning framework that unifies both. UniPred uses large language models (LLMs) to propose predicate effect distributions that supervise neural predicate learning from low-level data, while learned feedback iteratively refines the LLM hypotheses. Leveraging strong visual foundation model features, UniPred learns robust predicate classifiers in cluttered scenes. We further propose a predicate evaluation method that supports symbolic models beyond STRIPS assumptions. Across five simulated and one real-robot domains, UniPred achieves 2-4 times higher success rates than top-down methods and 3-4 times faster learning than bottom-up approaches, advancing scalable and flexible symbolic world modeling for robotics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_17992 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Unifying Deep Predicate Invention with Pre-trained Foundation Models Wang, Qianwei Li, Bowen Luo, Zhanpeng Xu, Yifan Gray, Alexander Silver, Tom Scherer, Sebastian Sycara, Katia Xie, Yaqi Robotics Long-horizon robotic tasks are hard due to continuous state-action spaces and sparse feedback. Symbolic world models help by decomposing tasks into discrete predicates that capture object properties and relations. Existing methods learn predicates either top-down, by prompting foundation models without data grounding, or bottom-up, from demonstrations without high-level priors. We introduce UniPred, a bilevel learning framework that unifies both. UniPred uses large language models (LLMs) to propose predicate effect distributions that supervise neural predicate learning from low-level data, while learned feedback iteratively refines the LLM hypotheses. Leveraging strong visual foundation model features, UniPred learns robust predicate classifiers in cluttered scenes. We further propose a predicate evaluation method that supports symbolic models beyond STRIPS assumptions. Across five simulated and one real-robot domains, UniPred achieves 2-4 times higher success rates than top-down methods and 3-4 times faster learning than bottom-up approaches, advancing scalable and flexible symbolic world modeling for robotics. |
| title | Unifying Deep Predicate Invention with Pre-trained Foundation Models |
| topic | Robotics |
| url | https://arxiv.org/abs/2512.17992 |