# Can Foundation Models Perform Zero-Shot Task Specification For Robot Manipulation?

arXiv: https://arxiv.org/abs/2204.11134 · HTML: https://mlsanity.com/paper/2204.11134/

## Key facts

- Cited by 70 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 46 references that are in the map.
- First posted on arXiv 23 Apr 2022.

## Builds on (most cited first, 30 of 37)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems](https://mlsanity.com/paper/2005.01643/) (2020)
- [Concrete Problems in AI Safety](https://mlsanity.com/paper/1606.06565/) (2016)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering](https://mlsanity.com/paper/2007.01282/) (2020)
- [Hindsight Experience Replay](https://mlsanity.com/paper/1707.01495/) (2017)
- [The "something something" video database for learning and evaluating visual common sense](https://mlsanity.com/paper/1706.04261/) (2017)
- [StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery](https://mlsanity.com/paper/2103.17249/) (2021)
- [Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations](https://mlsanity.com/paper/1709.10087/) (2017)
- [CLIPort: What and Where Pathways for Robotic Manipulation](https://mlsanity.com/paper/2109.12098/) (2021)
- [Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?](https://mlsanity.com/paper/2003.11539/) (2020)
- [Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning](https://mlsanity.com/paper/1910.11956/) (2019)
- [Pretrained Transformers for Text Ranking: BERT and Beyond](https://mlsanity.com/paper/2010.06467/) (2020)
- [Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observations](https://mlsanity.com/paper/1904.06387/) (2019)
- [Visual Reinforcement Learning with Imagined Goals](https://mlsanity.com/paper/1807.04742/) (2018)
- [Supersizing Self-supervision: Learning to Grasp from 50K Tries and 700 Robot Hours](https://mlsanity.com/paper/1509.06825/) (2015)
- [Simple but Effective: CLIP Embeddings for Embodied AI](https://mlsanity.com/paper/2111.09888/) (2021)
- [CLIPDraw: Exploring Text-to-Drawing Synthesis through Language-Image Encoders](https://mlsanity.com/paper/2106.14843/) (2021)
- [Provably efficient RL with Rich Observations via Latent State Decoding](https://mlsanity.com/paper/1901.09018/) (2019)
- [The Unsurprising Effectiveness of Pre-Trained Vision Models for Control](https://mlsanity.com/paper/2203.03580/) (2022)
- [End-to-End Robotic Reinforcement Learning without Reward Engineering](https://mlsanity.com/paper/1904.07854/) (2019)
- [Learning to Generalize Across Long-Horizon Tasks from Human Demonstrations](https://mlsanity.com/paper/2003.06085/) (2020)
- [C-Learning: Learning to Achieve Goals via Recursive Classification](https://mlsanity.com/paper/2011.08909/) (2020)
- [The IKEA ASM Dataset: Understanding People Assembling Furniture through Actions, Objects and Pose](https://mlsanity.com/paper/2007.00394/) (2020)
- [Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstration](https://mlsanity.com/paper/2105.06411/) (2021)

## Built on by (most cited first, 30 of 70)

- [Inner Monologue: Embodied Reasoning through Planning with Language Models](https://mlsanity.com/paper/2207.05608/) (2022)
- [VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models](https://mlsanity.com/paper/2307.05973/) (2023)
- [SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities](https://mlsanity.com/paper/2401.12168/) (2024)
- [VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Training](https://mlsanity.com/paper/2210.00030/) (2022)
- [PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs](https://mlsanity.com/paper/2402.07872/) (2024)
- [LIV: Language-Image Representations and Rewards for Robotic Control](https://mlsanity.com/paper/2306.00958/) (2023)
- [CREPE: Can Vision-Language Foundation Models Reason Compositionally?](https://mlsanity.com/paper/2212.07796/) (2022)
- [Language-Driven Representation Learning for Robotics](https://mlsanity.com/paper/2302.12766/) (2023)
- [Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning](https://mlsanity.com/paper/2310.12921/) (2023)
- [RL-VLM-F: Reinforcement Learning from Vision Language Foundation Model Feedback](https://mlsanity.com/paper/2402.03681/) (2024)
- [Vision-Language Models as Success Detectors](https://mlsanity.com/paper/2303.07280/) (2023)
- [Toward General-Purpose Robots via Foundation Models: A Survey and Meta-Analysis](https://mlsanity.com/paper/2312.08782/) (2023)
- [Motif: Intrinsic Motivation from Artificial Intelligence Feedback](https://mlsanity.com/paper/2310.00166/) (2023)
- [An Inverse Scaling Law for CLIP Training](https://mlsanity.com/paper/2305.07017/) (2023)
- [Unpacking Failure Modes of Generative Policies: Runtime Monitoring of Consistency and Progress](https://mlsanity.com/paper/2410.04640/) (2024)
- [Vision-Language Models as a Source of Rewards](https://mlsanity.com/paper/2312.09187/) (2023)
- [Real-World Robot Applications of Foundation Models: A Review](https://mlsanity.com/paper/2402.05741/) (2024)
- [Pre-Trained Image Encoder for Generalizable Visual Reinforcement Learning](https://mlsanity.com/paper/2212.08860/) (2022)
- [Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisons](https://mlsanity.com/paper/2603.02115/) (2026)
- [Retrospectives on the Embodied AI Workshop](https://mlsanity.com/paper/2210.06849/) (2022)
- [Learning Navigational Visual Representations with Semantic Map Supervision](https://mlsanity.com/paper/2307.12335/) (2023)
- [Language Reward Modulation for Pretraining Reinforcement Learning](https://mlsanity.com/paper/2308.12270/) (2023)
- [RT-Affordance: Affordances are Versatile Intermediate Representations for Robot Manipulation](https://mlsanity.com/paper/2411.02704/) (2024)
- [RT-Sketch: Goal-Conditioned Imitation Learning from Hand-Drawn Sketches](https://mlsanity.com/paper/2403.02709/) (2024)
- [Code as Reward: Empowering Reinforcement Learning with VLMs](https://mlsanity.com/paper/2402.04764/) (2024)
- [FlowRetrieval: Flow-Guided Data Retrieval for Few-Shot Imitation Learning](https://mlsanity.com/paper/2408.16944/) (2024)
- [Goal Representations for Instruction Following: A Semi-Supervised Language Interface to Control](https://mlsanity.com/paper/2307.00117/) (2023)
- [GenRL: Multimodal-foundation world models for generalization in embodied agents](https://mlsanity.com/paper/2406.18043/) (2024)
- [Task Success is not Enough: Investigating the Use of Video-Language Models as Behavior Critics for Catching Undesirable Agent Behaviors](https://mlsanity.com/paper/2402.04210/) (2024)
- [CLIPA-v2: Scaling CLIP Training with 81.1% Zero-shot ImageNet Accuracy within a 10,000 Budget; An Extra 4,000 Unlocks 81.8% Accuracy](https://mlsanity.com/paper/2306.15658/) (2023)

## Abstract (arXiv)

Task specification is at the core of programming autonomous robots. A low-effort modality for task specification is critical for engagement of non-expert end-users and ultimate adoption of personalized robot agents. A widely studied approach to task specification is through goals, using either compact state vectors or goal images from the same robot scene. The former is hard to interpret for non-experts and necessitates detailed state estimation and scene understanding. The latter requires the generation of desired goal image, which often requires a human to complete the task, defeating the purpose of having autonomous robots. In this work, we explore alternate and more general forms of goal specification that are expected to be easier for humans to specify and use such as images obtained from the internet, hand sketches that provide a visual description of the desired task, or simple language descriptions. As a preliminary step towards this, we investigate the capabilities of large scale pre-trained models (foundation models) for zero-shot goal specification, and find promising results in a collection of simulated robot manipulation tasks and real-world datasets.