# Robots That Take Initiative: A Framework for Building and Evaluating Proactive Robots

arXiv: https://arxiv.org/abs/2609.28910 · HTML: https://mlsanity.com/paper/2609.28910/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 70 references that are in the map.
- First posted on arXiv 24 Sep 2026.

## Builds on (most cited first, 30 of 36)

- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [Hindsight Experience Replay](https://mlsanity.com/paper/1707.01495/) (2017)
- [Behavioral Cloning from Observation](https://mlsanity.com/paper/1805.01954/) (2018)
- [VirtualHome: Simulating Household Activities via Programs](https://mlsanity.com/paper/1806.07011/) (2018)
- [Learning Latent Plans from Play](https://mlsanity.com/paper/1903.01973/) (2019)
- [Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners](https://mlsanity.com/paper/2307.01928/) (2023)
- [Watch-And-Help: A Challenge for Social Perception and Human-AI Collaboration](https://mlsanity.com/paper/2010.09890/) (2020)
- [Actionable Models: Unsupervised Offline Reinforcement Learning of Robotic Skills](https://mlsanity.com/paper/2104.07749/) (2021)
- [On Bringing Robots Home](https://mlsanity.com/paper/2311.16098/) (2023)
- [Learning by Watching: Physical Imitation of Manipulation Skills from Human Videos](https://mlsanity.com/paper/2101.07241/) (2021)
- [Bottom-Up Skill Discovery from Unsegmented Demonstrations for Long-Horizon Robot Manipulation](https://mlsanity.com/paper/2109.13841/) (2021)
- [RoboCasa365: A Large-Scale Simulation Framework for Training and Benchmarking Generalist Robots](https://mlsanity.com/paper/2603.04356/) (2026)
- [Online Bayesian Goal Inference for Boundedly-Rational Planning Agents](https://mlsanity.com/paper/2006.07532/) (2020)
- [Efficient Model Learning for Human-Robot Collaborative Tasks](https://mlsanity.com/paper/1405.6341/) (2014)
- [CLARA: Classifying and Disambiguating User Commands for Reliable Interactive Robotic Agents](https://mlsanity.com/paper/2306.10376/) (2023)
- [NOPA: Neurally-guided Online Probabilistic Assistance for Building Socially Intelligent Home Assistants](https://mlsanity.com/paper/2301.05223/) (2023)
- [Bridging Language and Action: A Survey of Language-Conditioned Robot Manipulation](https://mlsanity.com/paper/2312.10807/) (2023)
- [Landmark-Based Approaches for Goal Recognition as Planning](https://mlsanity.com/paper/1904.11739/) (2019)
- [Semantic Linking Maps for Active Visual Object Search](https://mlsanity.com/paper/2006.10807/) (2020)
- [Proactive Robot Assistance via Spatio-Temporal Object Modeling](https://mlsanity.com/paper/2211.15501/) (2022)
- [Coordination with Humans via Strategy Matching](https://mlsanity.com/paper/2210.15099/) (2022)
- [AssistanceZero: Scalably Solving Assistance Games](https://mlsanity.com/paper/2504.07091/) (2025)
- [LSTM-Based Goal Recognition in Latent Space](https://mlsanity.com/paper/1808.05249/) (2018)
- [Robi Butler: Multimodal Remote Interaction with a Household Robot Assistant](https://mlsanity.com/paper/2409.20548/) (2024)
- [ADAPT: Actively Discovering and Adapting to Preferences for any Task](https://mlsanity.com/paper/2504.04040/) (2025)
- [Predicting Routine Object Usage for Proactive Robot Assistance](https://mlsanity.com/paper/2309.06252/) (2023)
- [Anticipate & Act : Integrating LLMs and Classical Planning for Efficient Task Execution in Household Environments](https://mlsanity.com/paper/2502.02066/) (2025)
- [Predictive Spatio-Temporal Scene Graphs for Semi-Static Scenes](https://mlsanity.com/paper/2605.00121/) (2026)
- [GRACE: Generating Socially Appropriate Robot Actions Leveraging LLMs and Human Explanations](https://mlsanity.com/paper/2409.16879/) (2024)
- [A Model-Agnostic Approach for Semantically Driven Disambiguation in Human-Robot Interaction](https://mlsanity.com/paper/2409.17004/) (2024)

## Abstract (arXiv)

Effective robot assistance beyond narrow roles and repetitive tasks requires robots to be proactive - to decide what needs to be done rather than waiting to be told. While proactivity is increasingly explored, it lacks a unified formulation, and work in the domain is typically evaluated offline against static human models that cannot capture the effect of a robot's actions on the environment and the user's own behavior. We introduce a unified formalism for proactive robot assistance, organize it into three levels, and provide a framework to address the highest level of unprompted proactive assistance. We then show that offline evaluation overstates performance in this setting, and contribute a closed-loop evaluation with a human model that adapts to the robot. Finally, we present a method, GAP, that instantiates our framework, learning from passive observation to anticipate user goals and act. Under closed-loop evaluation, prior state-of-the-art methods collapse, in some cases adding more work than they save, while GAP remains robust and substantially outperforms them.