Tag
Introduces Object-Centric Environment Modeling (OCM), a method that organizes LLM agent experience into two executable code bases (object knowledge and procedure knowledge) to improve reuse, verification, and reduce invalid actions in interactive environments.
OPINE-World introduces an LLM agent that learns an object-centric programmatic world model online through interaction, using ontology-error-prioritized exploration and cooperating hypothesis-test agents, achieving strong results on ARC-AGI-3.
This paper proposes EHHN, an event-driven heterogeneous hypergraph network for object-centric next activity prediction in service processes, achieving state-of-the-art accuracy and memory efficiency on four benchmarks.
An object-centric residual reinforcement learning framework enhances zero-shot sim-to-real transfer for vision-language-action models, improving success rates from 42% to 76% on manipulation tasks without real-world training.
COMET is a model-based reinforcement learning algorithm that combines a frozen object-centric encoder with a transformer-based world model and Monte Carlo Tree Search, using causal attention to focus on task-relevant objects, achieving higher scores on visual RL benchmarks.