# Offline Discovery of Interpretable Skills from Multi-Task Trajectories

arXiv: https://arxiv.org/abs/2602.01018 · HTML: https://mlsanity.com/paper/2602.01018/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 22 of its 26 references that are in the map.
- First posted on arXiv 1 Feb 2026.

## Placed in

- Offline RL and imitation learning from data / [Robot manipulation imitation learning](https://mlsanity.com/offlinerl/robot-manipulation-imitation-learning/)

## Builds on (most cited first, 22 of 22)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems](https://mlsanity.com/paper/2005.01643/) (2020)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [D4RL: Datasets for Deep Data-Driven Reinforcement Learning](https://mlsanity.com/paper/2004.07219/) (2020)
- [A Generalist Agent](https://mlsanity.com/paper/2205.06175/) (2022)
- [Diversity is All You Need: Learning Skills without a Reward Function](https://mlsanity.com/paper/1802.06070/) (2018)
- [Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning](https://mlsanity.com/paper/1910.11956/) (2019)
- [Dynamics-Aware Unsupervised Discovery of Skills](https://mlsanity.com/paper/1907.01657/) (2019)
- [OPAL: Offline Primitive Discovery for Accelerating Offline Reinforcement Learning](https://mlsanity.com/paper/2010.13611/) (2020)
- [A Survey of Imitation Learning: Algorithms, Recent Developments, and Challenges](https://mlsanity.com/paper/2309.02473/) (2023)
- [A Survey on Vision-Language-Action Models: An Action Tokenization Perspective](https://mlsanity.com/paper/2507.01925/) (2025)
- [Diffusion-VLA: Generalizable and Interpretable Robot Foundation Model via Self-Generated Reasoning](https://mlsanity.com/paper/2412.03293/) (2024)
- [Reasoning with Latent Diffusion in Offline Reinforcement Learning](https://mlsanity.com/paper/2309.06599/) (2023)
- [Full-Order Sampling-Based MPC for Torque-Level Locomotion Control via Diffusion-Style Annealing](https://mlsanity.com/paper/2409.15610/) (2024)
- [Deep Generative Models for Offline Policy Learning: Tutorial, Survey, and Perspectives on Future Directions](https://mlsanity.com/paper/2402.13777/) (2024)
- [Multi-task Hierarchical Adversarial Inverse Reinforcement Learning](https://mlsanity.com/paper/2305.12633/) (2023)
- [DexSkills: Skill Segmentation Using Haptic Data for Learning Autonomous Long-Horizon Robotic Manipulation Tasks](https://mlsanity.com/paper/2405.03476/) (2024)
- [Radial basis function kernel optimization for Support Vector Machine classifiers](https://mlsanity.com/paper/2007.08233/) (2020)
- [A Unified Algorithm Framework for Unsupervised Discovery of Skills based on Determinantal Point Process](https://mlsanity.com/paper/2212.00211/) (2022)
- [SkillS: Adaptive Skill Sequencing for Efficient Temporally-Extended Exploration](https://mlsanity.com/paper/2211.13743/) (2022)
- [Self-Calibrating Anomaly and Change Detection for Autonomous Inspection Robots](https://mlsanity.com/paper/2209.02379/) (2022)

## Built on by (most cited first, 1 of 1)

- [RELIC: Revealed Principles for Learning Interpretable Composable Skills in Multi-Agent Planning](https://mlsanity.com/paper/2607.16745/) (2026)

## Abstract (arXiv)

Hierarchical Imitation Learning is a powerful paradigm for acquiring complex robot behaviors from demonstrations. A central challenge, however, lies in discovering reusable skills from long-horizon, multi-task offline data, especially when the data lacks explicit rewards or subtask annotations. In this work, we introduce LOKI, a three-stage end-to-end learning framework designed for offline skill discovery and hierarchical imitation. The framework commences with a two-stage, weakly supervised skill discovery process: Stage one performs coarse, task-aware macro-segmentation by employing an alignment-enforced Vector Quantized VAE guided by weak task labels. Stage two then refines these segments at a micro-level using a self-supervised sequential model, followed by an iterative clustering process to consolidate skill boundaries. The third stage then leverages these precise boundaries to construct a hierarchical policy within an option-based framework-complete with a learned termination condition beta for explicit skill switching. LOKI achieves high success rates on the challenging D4RL Kitchen benchmark and outperforms standard HIL baselines. Furthermore, we demonstrate that the discovered skills are semantically meaningful, aligning with human intuition, and exhibit compositionality by successfully sequencing them to solve a novel, unseen task.