Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ADaPT: As-Needed Decomposition and Planning with Language Models

About

Large Language Models (LLMs) are increasingly being used for interactive decision-making tasks requiring planning and adapting to the environment. Recent works employ LLMs-as-agents in broadly two ways: iteratively determining the next action (iterative executors) or generating plans and executing sub-tasks using LLMs (plan-and-execute). However, these methods struggle with task complexity, as the inability to execute any sub-task may lead to task failure. To address these shortcomings, we introduce As-Needed Decomposition and Planning for complex Tasks (ADaPT), an approach that explicitly plans and decomposes complex sub-tasks as-needed, i.e., when the LLM is unable to execute them. ADaPT recursively decomposes sub-tasks to adapt to both task complexity and LLM capability. Our results demonstrate that ADaPT substantially outperforms established strong baselines, achieving success rates up to 28.3% higher in ALFWorld, 27% in WebShop, and 33% in TextCraft -- a novel compositional dataset that we introduce. Through extensive analysis, we illustrate the importance of multilevel decomposition and establish that ADaPT dynamically adjusts to the capabilities of the executor LLM as well as to task complexity.

Archiki Prasad, Alexander Koller, Mareike Hartmann, Peter Clark, Ashish Sabharwal, Mohit Bansal, Tushar Khot• 2023

Related benchmarks

TaskDatasetResultRank
Interactive Decision-makingAlfWorld
Overall Success Rate16.67
398
Interactive Decision-makingScienceWorld
Success Rate26.2
78
Interactive Decision-makingWebshop
Success Rate32
77
Interactive Decision-makingTextCraft
Success Rate30.2
60
Agent TaskWebshop
Success Rate32
57
Item craftingTextCraft (test)
Success Rate58
32
Embodied agentAlfWorld
Success Rate72.39
31
Embodied InteractionAlfWorld
Success Rate72.39
22
Interactive Decision-makingALFWorld OOD
Success Rate10.26
18
Language Agent TaskTextCraft
Success Rate (SR)77
17
Showing 10 of 15 rows

Other info

Follow for update