# MIND: Instruction-Conditioned Incremental Grounding for Robots

arXiv: https://arxiv.org/abs/2610.19262 · HTML: https://mlsanity.com/paper/2610.19262/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 30 references that are in the map.
- First posted on arXiv 13 Oct 2026.

## Builds on (most cited first, 24 of 24)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [Inner Monologue: Embodied Reasoning through Planning with Language Models](https://mlsanity.com/paper/2207.05608/) (2022)
- [AI2-THOR: An Interactive 3D Environment for Visual AI](https://mlsanity.com/paper/1712.05474/) (2017)
- [ProcTHOR: Large-Scale Embodied AI Using Procedural Generation](https://mlsanity.com/paper/2206.06994/) (2022)
- [Scaling Open-Vocabulary Object Detection](https://mlsanity.com/paper/2306.09683/) (2023)
- [Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners](https://mlsanity.com/paper/2307.01928/) (2023)
- [SayPlan: Grounding Large Language Models using 3D Scene Graphs for Scalable Robot Task Planning](https://mlsanity.com/paper/2307.06135/) (2023)
- [Text2Motion: From Natural Language Instructions to Feasible Plans](https://mlsanity.com/paper/2303.12153/) (2023)
- [Grounded Decoding: Guiding Text Generation with Grounded Models for Embodied Agents](https://mlsanity.com/paper/2303.00855/) (2023)
- [CLARA: Classifying and Disambiguating User Commands for Reliable Interactive Robotic Agents](https://mlsanity.com/paper/2306.10376/) (2023)
- [LiP-LLM: Integrating Linear Programming and dependency graph with Large Language Models for multi-robot task planning](https://mlsanity.com/paper/2410.21040/) (2024)
- [DART-LLM: Dependency-Aware Multi-Robot Task Decomposition and Execution using Large Language Models](https://mlsanity.com/paper/2411.09022/) (2024)
- [DeComplex: Task planning from complex natural instructions by a collocating robot](https://mlsanity.com/paper/2008.10084/) (2020)
- [LINGO-Space: Language-Conditioned Incremental Grounding for Space](https://mlsanity.com/paper/2402.01183/) (2024)
- [STMA: A Spatio-Temporal Memory Agent for Long-Horizon Embodied Task Planning](https://mlsanity.com/paper/2502.10177/) (2025)
- [KEEP: A KV-Cache-Centric Memory Management System for Efficient Embodied Planning](https://mlsanity.com/paper/2602.23592/) (2026)
- [DynaCon: Dynamic Robot Planner with Contextual Awareness via LLMs](https://mlsanity.com/paper/2309.16031/) (2023)
- [AgenticCache: Cache-Driven Asynchronous Planning for Embodied AI Agents](https://mlsanity.com/paper/2604.24039/) (2026)
- [Context Matters! Relaxing Goals with LLMs for Feasible 3D Scene Planning](https://mlsanity.com/paper/2506.15828/) (2025)
- [Event-Grounding Graph: Unified Spatio-Temporal Scene Graph from Robotic Observations](https://mlsanity.com/paper/2510.18697/) (2025)
- [Integrated Exploration and Sequential Manipulation on Scene Graph with LLM-based Situated Replanning](https://mlsanity.com/paper/2602.04419/) (2026)
- [When Memory Lies: An Empirical Study of Spatial Memory Staleness in VLM Agents](https://mlsanity.com/paper/2608.04574/) (2026)
- [Mimir: A Neuro-Symbolic Memory System with Dynamic Grounding for Embodied Agents in Interactive Environments](https://mlsanity.com/paper/2608.04933/) (2026)

## Abstract (arXiv)

Robots following persistent language instructions must decide which cached room judgments remain valid as their environment changes. We present MIND (Maintenance via INstruction-conditioned Dependencies), a framework for maintaining target-room decisions without re-evaluating every room after each event. A language model compiles each instruction into dependencies for semantic qualification and target selection. The controller intersects semantic dependencies with event footprints, refreshes affected judgments, and updates reachability and selection separately. For fixed deterministic evaluation, planning, and selection, conservative dependency and footprint coverage preserves the decisions of full recomputation. On 240 five-event ProcTHOR sequences with an exact benchmark predicate, MIND reduces mean semantic evaluations from 30.6 to 11.2 while preserving all 1,200 post-event decisions. Frozen-evaluator experiments with Phi-4 and Qwen3-14B also preserve all 1,200 decisions, without improving the evaluators' underlying task accuracy. In 32 closed-loop AI2-THOR episodes using exact-predicate qualification and metadata-derived footprints, MIND matches the final target decision of full recomputation in every episode. Unitree G1 experiments demonstrate target reuse and reselection across room transitions without additional semantic evaluations under MIND, and abstention after a barrier and an object removal at 4 evaluations against 15 to 18. Additional materials are available at https://mind-icra27.github.io/.