mlsanityData to 7 Oct 2026

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

arXiv 2602.05048 ↗First posted 4 Feb 2026cs.AI cs.HCcited by 4

On the map: LLM reasoning → Reasoning for agents, planning and code

Abstract (arXiv)

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.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 35 of its 63 references that are in the map. First posted on arXiv 4 Feb 2026.

Builds on

35 of 63 references.

References · 35

  1. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  2. 2017Deep reinforcement learning from human preferencescited by 3,846
  3. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  4. 2022Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agentscited by 1,029
  5. 2023BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Datasetcited by 828
  6. 2024TravelPlanner: A Benchmark for Real-World Planning with Language Agentscited by 385
  7. 2019On the Utility of Learning about Humans for Human-AI Coordinationcited by 345
  8. 2023Minigrid & Miniworld: Modular & Customizable Reinforcement Learning Environments for Goal-Oriented Taskscited by 320
  9. 2018Programmatically Interpretable Reinforcement Learningcited by 192
  10. 2017Visualizing and Understanding Atari Agentscited by 141
  11. 2023SmartPlay: A Benchmark for LLMs as Intelligent Agentscited by 123
  12. 2021Recurrent Model-Free RL Can Be a Strong Baseline for Many POMDPscited by 115
23 more
  1. 2021Why Generalization in RL is Difficult: Epistemic POMDPs and Implicit Partial Observabilitycited by 102
  2. 2018SDRL: Interpretable and Data-efficient Deep Reinforcement Learning Leveraging Symbolic Planningcited by 77
  3. 2022Help me write a poem: Instruction Tuning as a Vehicle for Collaborative Poetry Writingcited by 71
  4. 2019Estimating Risk and Uncertainty in Deep Reinforcement Learningcited by 68
  5. 2021Memory-based Deep Reinforcement Learning for POMDPscited by 53
  6. 2022A Review of Uncertainty for Deep Reinforcement Learningcited by 36
  7. 2023Think, Act, and Ask: Open-World Interactive Personalized Robot Navigationcited by 31
  8. 2023LLM A*: Human in the Loop Large Language Models Enabled A* Search for Roboticscited by 26
  9. 2020Lipschitz Lifelong Reinforcement Learningcited by 23
  10. 2024End-to-End Neuro-Symbolic Reinforcement Learning with Textual Explanationscited by 17
  11. 2023Language and Sketching: An LLM-driven Interactive Multimodal Multitask Robot Navigation Frameworkcited by 16
  12. 2025InSPO: Unlocking Intrinsic Self-Reflection for LLM Preference Optimizationcited by 11
  13. 2022Asking for Knowledge: Training RL Agents to Query External Knowledge Using Languagecited by 10
  14. 2022Symbolic Visual Reinforcement Learning: A Scalable Framework with Object-Level Abstraction and Differentiable Expression Searchcited by 9
  15. 2023Decision Making for Human-in-the-loop Robotic Agents via Uncertainty-Aware Reinforcement Learningcited by 9
  16. 2024Learning from Random Demonstrations: Offline Reinforcement Learning with Importance-Sampled Diffusion Modelscited by 8
  17. 2025Lipschitz Lifelong Monte Carlo Tree Search for Mastering Non-Stationary Taskscited by 8
  18. 2025A Neurosymbolic Framework for Interpretable Cognitive Attack Detection in Augmented Realitycited by 8
  19. 2025Calibrating and Rotating: A Unified Framework for Weight Conditioning in PEFTcited by 4
  20. 2023Learning When to Ask for Help: Efficient Interactive Navigation via Implicit Uncertainty Estimationcited by 3
  21. 2024RGMDT: Return-Gap-Minimizing Decision Tree Extraction in Non-Euclidean Metric Spacecited by 2
  22. 2025Tunable Leg Stiffness in a Monopedal Hopper for Energy-Efficient Vertical Hopping Across Varying Ground Profilescited by 2
  23. 2026ACDZero: MCTS Agent for Mastering Automated Cyber Defensecited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026NonZero: Interaction-Guided Exploration for Multi-Agent Monte Carlo Tree Searchcited by 4
  2. 2026Reasoning Knowledge-Gap in Drone Planning via LLM-based Active Elicitationcited by 1
  3. 2026TIPS: Topological Ill-Posedness Probing and Steering in Large Language Modelscited by 0
  4. 2026FlowEdit: Information-Theoretic Control of LLM Reasoning Flows for Ill-posed Problems Involving Conflictscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025LOOP: A Plug-and-Play Neuro-Symbolic Framework for Enhancing Planning in Autonomous Systems
  2. 2025Closing the Expression Gap in LLM Instructions via Socratic Questioning
  3. 2025Emergence: Overcoming Privileged Information Bias in Asymmetric Embodied Agents via Active Querying
  4. 2025Hierarchical Planning for Complex Tasks with Knowledge Graph-RAG and Symbolic Verification
  5. 2026Bilevel Planning with Learned Symbolic Abstractions from Interaction Data ↗
  6. 2026DUPLEX: Agentic Dual-System Planning via LLM-Driven Information Extraction
  7. 2026Extracting Search Trees from LLM Reasoning Traces Reveals Myopic Planning
  8. 2024Revealing the Barriers of Language Agents in Planning