# GRAIL: Goal Recognition Alignment through Imitation Learning

arXiv: https://arxiv.org/abs/2602.14252 · HTML: https://mlsanity.com/paper/2602.14252/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 20 of its 56 references that are in the map.
- First posted on arXiv 15 Feb 2026.

## Placed in

- Offline RL and imitation learning from data / [Imitation learning methods](https://mlsanity.com/offlinerl/imitation-learning-methods/)

## Builds on (most cited first, 20 of 20)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [A Survey of Inverse Reinforcement Learning: Challenges, Methods and Progress](https://mlsanity.com/paper/1806.06877/) (2018)
- [The Complexity of Decentralized Control of Markov Decision Processes](https://mlsanity.com/paper/1301.3836/) (2013)
- [Consequences of Misaligned AI](https://mlsanity.com/paper/2102.03896/) (2021)
- [Online Bayesian Goal Inference for Boundedly-Rational Planning Agents](https://mlsanity.com/paper/2006.07532/) (2020)
- [panda-gym: Open-source goal-conditioned environments for robotic learning](https://mlsanity.com/paper/2106.13687/) (2021)
- [Humans are not Boltzmann Distributions: Challenges and Opportunities for Modelling Human Feedback and Interaction in Reinforcement Learning](https://mlsanity.com/paper/2206.13316/) (2022)
- [Landmark-Based Approaches for Goal Recognition as Planning](https://mlsanity.com/paper/1904.11739/) (2019)
- [Active Goal Recognition](https://mlsanity.com/paper/1909.11173/) (2019)
- [Multimodal Uncertainty Reduction for Intention Recognition in Human-Robot Interaction](https://mlsanity.com/paper/1907.02426/) (2019)
- [Goal Recognition as Reinforcement Learning](https://mlsanity.com/paper/2202.06356/) (2022)
- [Goal Recognition as a Deep Learning Task: the GRNet Approach](https://mlsanity.com/paper/2210.02377/) (2022)
- [Norm Conflict Resolution in Stochastic Domains](https://mlsanity.com/paper/1706.07448/) (2017)
- [An LP-Based Approach for Goal Recognition as Planning](https://mlsanity.com/paper/1905.04210/) (2019)
- [ODGR: Online Dynamic Goal Recognition](https://mlsanity.com/paper/2407.16220/) (2024)
- [Goal Recognition using Actor-Critic Optimization](https://mlsanity.com/paper/2501.01463/) (2024)
- [Artificial Intelligent Disobedience: Rethinking the Agency of Our Artificial Teammates](https://mlsanity.com/paper/2506.22276/) (2025)
- [SLIM: Semi-Lazy Inference Mechanism for Plan Recognition](https://mlsanity.com/paper/1703.00838/) (2017)
- [A unified energy optimality criterion predicts human navigation paths and speeds](https://mlsanity.com/paper/2001.02287/) (2020)

## Built on by (most cited first, 3 of 3)

- [Beyond Mimicry: Toward Lifelong Adaptability in Imitation Learning](https://mlsanity.com/paper/2602.19930/) (2026)
- [Online Goal Recognition using Path Signature and Dynamic Time Warping](https://mlsanity.com/paper/2605.07736/) (2026)
- [Improving Online Reinforcement Learning via Bidirectional Behavior Prior Distillation](https://mlsanity.com/paper/2609.20268/) (2026)

## Abstract (arXiv)

Understanding an agent's goals from its behavior is fundamental to aligning AI systems with human intentions. Existing goal recognition methods typically rely on an optimal goal-oriented policy representation, which may differ from the actor's true behavior and hinder the accurate recognition of their goal. To address this gap, this paper introduces Goal Recognition Alignment through Imitation Learning (GRAIL), which leverages imitation learning and inverse reinforcement learning to learn one goal-directed policy for each candidate goal directly from (potentially suboptimal) demonstration trajectories. By scoring an observed partial trajectory with each learned goal-directed policy in a single forward pass, GRAIL retains the one-shot inference capability of classical goal recognition while leveraging learned policies that can capture suboptimal and systematically biased behavior. Across the evaluated domains, GRAIL increases the F1-score by more than 0.5 under systematically biased optimal behavior, achieves gains of approximately 0.1-0.3 under suboptimal behavior, and yields improvements of up to 0.4 under noisy optimal trajectories, while remaining competitive in fully optimal settings. This work contributes toward scalable and robust models for interpreting agent goals in uncertain environments.