MINT: Minimal Information Neuro-Symbolic Tree for Objective-Driven Knowledge-Gap Reasoning and Active Elicitation

2026-02-04

AI Summary

This work addresses the knowledge gap arising from missing goals and object information in open-world human-AI collaborative planning, which often impedes effective joint decision-making. To bridge this gap, the paper proposes MINT, a novel framework that integrates neuro-symbolic trees with active knowledge acquisition for the first time. MINT leverages neuro-symbolic reasoning to identify knowledge gaps and employs self-play reinforcement learning to optimize the AI’s questioning strategy, enabling it to elicit critical human input with minimal interaction. The approach combines large language model–assisted reasoning and query generation within an extended Markov decision process formulation. Evaluated on three benchmark tasks involving unknown objects, MINT achieves near-expert-level returns with only a few queries, significantly improving both task success rates and cumulative rewards.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Multiagent Systems: Multiagent Planning
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation

Abstract

Joint planning through language-based interactions is a key area of human-AI teaming. Planning problems in the open world often involve various aspects of incomplete information and unknowns, e.g., objects involved, human goals/intents -- thus leading to knowledge gaps in joint planning. We consider the problem of discovering optimal interaction strategies for AI agents to actively elicit human inputs in object-driven planning. To this end, we propose Minimal Information Neuro-Symbolic Tree (MINT) to reason about the impact of knowledge gaps and leverage self-play with MINT to optimize the AI agent's elicitation strategies and queries. More precisely, MINT builds a symbolic tree by making propositions of possible human-AI interactions and by consulting a neural planning policy to estimate the uncertainty in planning outcomes caused by remaining knowledge gaps. Finally, we leverage LLM to search and summarize MINT's reasoning process and curate a set of queries to optimally elicit human inputs for best planning performance. By considering a family of extended Markov decision processes with knowledge gaps, we analyze the return guarantee for a given MINT with active human elicitation. Our evaluation on three benchmarks involving unseen/unknown objects of increasing realism shows that MINT-based planning attains near-expert returns by issuing a limited number of questions per task while achieving significantly improved rewards and success rates.

Problem

• knowledge gaps
• active elicitation
• human-AI teaming
• joint planning
• incomplete information

Innovation

• Neuro-Symbolic Reasoning
• Active Elicitation
• Knowledge-Gap Reasoning
• Human-AI Teaming
• LLM-Augmented Planning