Abstract:A typical way in which a machine acquires knowledge from humans is by programming. Compared to learning from demonstrations or experiences, programmatic learning allows the machine to acquire a novel skill as soon as the program is written, and, by building a library of programs, a machine can quickly learn how to perform complex tasks. However, as programs often take their execution contexts for granted, they are brittle when the contexts change, making it difficult to adapt complex programs to new contexts. We present natural programming, a library learning system that combines programmatic learning with a hierarchical planner. Natural programming maintains a library of decompositions, consisting of a goal, a linguistic description of how this goal decompose into sub-goals, and a concrete instance of its decomposition into sub-goals. A user teaches the system via curriculum building, by identifying a challenging yet not impossible goal along with linguistic hints on how this goal may be decomposed into sub-goals. The system solves for the goal via hierarchical planning, using the linguistic hints to guide its probability distribution in proposing the right plans. The system learns from this interaction by adding newly found decompositions in the successful search into its library. Simulated studies and a human experiment (n=360) on a controlled environment demonstrate that natural programming can robustly compose programs learned from different users and contexts, adapting faster and solving more complex tasks when compared to programmatic baselines.

Learning to Order Facts for Discourse Planning in Natural Language Generation

Learning to Plan with Natural Language

Learning to Plan by Updating Natural Language

Can Neural Generators for Dialogue Learn Sentence Planning and Discourse Structuring?

Learning adaptive planning representations with natural language guidance

Learning to Plan for Language Modeling from Unlabeled Data

Learning to Order Natural Language Texts.

Learning to reason over scene graphs: a case study of finetuning GPT-2 into a robot language model for grounded task planning

Large Language Models as Planning Domain Generators

Planning in the Dark: LLM-Symbolic Planning Pipeline without Experts

Translating Natural Language to Planning Goals with Large-Language Models

NATURAL PLAN: Benchmarking LLMs on Natural Language Planning

Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents

Deep Learning for Generalised Planning with Background Knowledge

What's the Plan? Evaluating and Developing Planning-Aware Techniques for Language Models

Learning a Hierarchical Planner from Humans in Multiple Generations

Learning to Program with Natural Language

Can Graph Learning Improve Planning in LLM-based Agents?

Learning to Plan for Retrieval-Augmented Large Language Models from Knowledge Graphs