Research map / Multi-agent systems and games / Human-AI and goal-conditioned
ExoPredicator: Learning Abstract Models of Dynamic Worlds for Robot Planning
Yichao Liang, Thanh Dat Nguyen, Cambridge Yang, Tianyang Li, Joshua B. Tenenbaum, Carl Edward Rasmussen, Adrian Weller, Zenna Tavares, Tom Silver, Kevin Ellis
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| learn abstract causal-process models instead of HRL (MAPLE) | ExoPredicator · 5 simulated tabletop environments (held-out tasks) | solve rate | Improved |
| learn causal processes instead of prompting a VLM zero-shot (ViLA) | ExoPredicator · 5 simulated tabletop environments (held-out tasks) | solve rate | Improved |
| add demonstrations to VLM planner (few-shot ViLA) | ViLA · 5 simulated tabletop environments (held-out tasks) | solve rate | Mixed |
| learn causal processes with delays instead of STRIPS-style operators (VisPred) | ExoPredicator · 5 simulated tabletop environments (held-out tasks) | solve rate | Improved |
| learn abstractions and parameters instead of manual abstractions | ExoPredicator · Grow and Domino | solve rate | Improved |
| learn delay parameters via variational inference instead of a single fixed delay (ablation) | ExoPredicator (Manual-d ablation) · 5 simulated tabletop environments (held-out tasks) | solve rate | Improved |
| add predicate invention and learning versus initial abstractions only (ablation) | ExoPredicator (No invent ablation) · 5 simulated tabletop environments (held-out tasks) | solve rate | Improved |
| add Bayesian model selection to LLM-proposed processes (ablation) | ExoPredicator (No Bayes ablation) · 5 simulated tabletop environments (held-out tasks) | solve rate | Improved |
| add LLM guidance for condition proposals instead of a heuristic (ablation) | ExoPredicator (No LLM ablation) · 5 simulated tabletop environments (held-out tasks) | solve rate | Improved |
Closest papers by meaning
- VisualPredicator: Learning Abstract World Models with Neuro-Symbolic Predicates for Robot Planning (ICLR 2025 · similarity 0.87)
- Curious Causality-Seeking Agents Learn Meta Causal World (NeurIPS 2025 · similarity 0.87)
- Modeling Unseen Environments with Language-guided Composable Causal Components in Reinforcement Learning (ICLR 2025 · similarity 0.86)
- Learning 3D Persistent Embodied World Models (NeurIPS 2025 · similarity 0.86)
- Embodied Task Planning via Graph-Informed Action Generation with Large Language Models (ICML 2026 · similarity 0.86)
- Dream to Manipulate: Compositional World Models Empowering Robot Imitation Learning with Imagination (ICLR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
