TL;DR
This work introduces tentative object pruning, an approach that leverages this property by constructing multiple simplified tasks with reduced object sets and searching them in parallel until a valid, satisficing solution is found.
This work proposes to abstract the planning task for abstract solutions that are then repaired into concrete plans for the input task, and showcases domains in which drawing an abstract plan and repairing it is significantly faster than solving the original problem.
GATS (Graph-Augmented Tree Search), a planning framework that combines systematic UCB1-based tree search with a layered world model to eliminate LLM calls during inference while achieving superior planning performance, is presented.
This work reframe trajectory evaluation as a distance between the agent's execution graph and a set of valid solution graphs, and instantiates OTAP via an unbalanced fused Gromov-Wasserstein transport problem over attributed dependency graphs, which is a pseudo-metric that is provably invariant to dependency-preserving...
Evidence Acquisition and Feasibility Gating (EAFG) is proposed, a framework that acquires visual evidence through VLM-generated exploratory subgoals and TAMP-based execution and applies a feasibility gate to decide whether to proceed with task planning, acquire further evidence, or halt.
UniETP is a unified interface integrating four commonly-used simulators integrating four commonly-used simulators that enhances task diversity and complexity across dimensions like task logic, instance grounding, and instruction understanding, constructing a new dataset with varied levels of difficulty in an automatic...
It is shown a proof-of-the-concept planner can sometimes solve the bounded proper planning problem by using a domain-independent heuristic that guides search over sequences of actions.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
