# A Minimalist Prompt for Zero-Shot Policy Learning

arXiv: https://arxiv.org/abs/2405.06063 · HTML: https://mlsanity.com/paper/2405.06063/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 45 references that are in the map.
- First posted on arXiv 9 May 2024.

## Builds on (most cited first, 30 of 40)

- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [PaLM-E: An Embodied Multimodal Language Model](https://mlsanity.com/paper/2303.03378/) (2023)
- [Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems](https://mlsanity.com/paper/2005.01643/) (2020)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?](https://mlsanity.com/paper/2202.12837/) (2022)
- [D4RL: Datasets for Deep Data-Driven Reinforcement Learning](https://mlsanity.com/paper/2004.07219/) (2020)
- [Soft Actor-Critic Algorithms and Applications](https://mlsanity.com/paper/1812.05905/) (2018)
- [Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning](https://mlsanity.com/paper/1910.10897/) (2019)
- [Hindsight Experience Replay](https://mlsanity.com/paper/1707.01495/) (2017)
- [A Generalist Agent](https://mlsanity.com/paper/2205.06175/) (2022)
- [ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks](https://mlsanity.com/paper/1912.01734/) (2019)
- [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://mlsanity.com/paper/2106.02039/) (2021)
- [BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning](https://mlsanity.com/paper/2202.02005/) (2022)
- [End-to-end Driving via Conditional Imitation Learning](https://mlsanity.com/paper/1710.02410/) (2017)
- [Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables](https://mlsanity.com/paper/1903.08254/) (2019)
- [One-Shot Imitation Learning](https://mlsanity.com/paper/1703.07326/) (2017)
- [Contextual Markov Decision Processes](https://mlsanity.com/paper/1502.02259/) (2015)
- [Multi-Game Decision Transformers](https://mlsanity.com/paper/2205.15241/) (2022)
- [Online Decision Transformer](https://mlsanity.com/paper/2202.05607/) (2022)
- [RvS: What is Essential for Offline RL via Supervised Learning?](https://mlsanity.com/paper/2112.10751/) (2021)
- [Goal-conditioned Imitation Learning](https://mlsanity.com/paper/1906.05838/) (2019)
- [DARLA: Improving Zero-Shot Transfer in Reinforcement Learning](https://mlsanity.com/paper/1707.08475/) (2017)
- [Prompting Decision Transformer for Few-Shot Policy Generalization](https://mlsanity.com/paper/2206.13499/) (2022)
- [Reinforcement Learning Upside Down: Don't Predict Rewards -- Just Map Them to Actions](https://mlsanity.com/paper/1912.02875/) (2019)
- [Zero-Shot Visual Imitation](https://mlsanity.com/paper/1804.08606/) (2018)
- [Learning to Generalize Across Long-Horizon Tasks from Human Demonstrations](https://mlsanity.com/paper/2003.06085/) (2020)

## Built on by (most cited first, 2 of 2)

- [The Judge Variable: Challenging Judge-Agnostic Legal Judgment Prediction](https://mlsanity.com/paper/2507.13732/) (2025)
- [Policy-as-Skill: Governed LLM Decision Support with Evidence, Deterministic Control, and Audit](https://mlsanity.com/paper/2609.27087/) (2026)

## Abstract (arXiv)

Transformer-based methods have exhibited significant generalization ability when prompted with target-domain demonstrations or example solutions during inference. Although demonstrations, as a way of task specification, can capture rich information that may be hard to specify by language, it remains unclear what information is extracted from the demonstrations to help generalization. Moreover, assuming access to demonstrations of an unseen task is impractical or unreasonable in many real-world scenarios, especially in robotics applications. These questions motivate us to explore what the minimally sufficient prompt could be to elicit the same level of generalization ability as the demonstrations. We study this problem in the contextural RL setting which allows for quantitative measurement of generalization and is commonly adopted by meta-RL and multi-task RL benchmarks. In this setting, the training and test Markov Decision Processes (MDPs) only differ in certain properties, which we refer to as task parameters. We show that conditioning a decision transformer on these task parameters alone can enable zero-shot generalization on par with or better than its demonstration-conditioned counterpart. This suggests that task parameters are essential for the generalization and DT models are trying to recover it from the demonstration prompt. To extract the remaining generalizable information from the supervision, we introduce an additional learnable prompt which is demonstrated to further boost zero-shot generalization across a range of robotic control, manipulation, and navigation benchmark tasks.