# Theia: Distilling Diverse Vision Foundation Models for Robot Learning

arXiv: https://arxiv.org/abs/2407.20179 · HTML: https://mlsanity.com/paper/2407.20179/

## Key facts

- Cited by 61 papers in the map (data through 14 Oct 2026).
- Builds on 68 of its 74 references that are in the map.
- First posted on arXiv 29 Jul 2024.

## Builds on (most cited first, 30 of 68)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Unsupervised Learning of Visual Features by Contrasting Cluster Assignments](https://mlsanity.com/paper/2006.09882/) (2020)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [Ego4D: Around the World in 3,000 Hours of Egocentric Video](https://mlsanity.com/paper/2110.07058/) (2021)
- [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://mlsanity.com/paper/2401.10891/) (2024)
- [Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning](https://mlsanity.com/paper/1910.10897/) (2019)
- [Habitat: A Platform for Embodied AI Research](https://mlsanity.com/paper/1904.01201/) (2019)
- [The "something something" video database for learning and evaluating visual common sense](https://mlsanity.com/paper/1706.04261/) (2017)
- [DeepMind Control Suite](https://mlsanity.com/paper/1801.00690/) (2018)
- [Open-vocabulary Object Detection via Vision and Language Knowledge Distillation](https://mlsanity.com/paper/2104.13921/) (2021)
- [Object-Centric Learning with Slot Attention](https://mlsanity.com/paper/2006.15055/) (2020)
- [Vision Transformers Need Registers](https://mlsanity.com/paper/2309.16588/) (2023)
- [Open X-Embodiment: Robotic Learning Datasets and RT-X Models](https://mlsanity.com/paper/2310.08864/) (2023)
- [Scaling Egocentric Vision: The EPIC-KITCHENS Dataset](https://mlsanity.com/paper/1804.02748/) (2018)
- [CURL: Contrastive Unsupervised Representations for Reinforcement Learning](https://mlsanity.com/paper/2004.04136/) (2020)

## Built on by (most cited first, 30 of 61)

- [Perception Encoder: The best visual embeddings are not at the output of the network](https://mlsanity.com/paper/2504.13181/) (2025)
- [DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge](https://mlsanity.com/paper/2507.04447/) (2025)
- [RADIOv2.5: Improved Baselines for Agglomerative Vision Foundation Models](https://mlsanity.com/paper/2412.07679/) (2024)
- [Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulation](https://mlsanity.com/paper/2411.18623/) (2024)
- [DUNE: Distilling a Universal Encoder from Heterogeneous 2D and 3D Teachers](https://mlsanity.com/paper/2503.14405/) (2025)
- [SPA: 3D Spatial-Awareness Enables Effective Embodied Representation](https://mlsanity.com/paper/2410.08208/) (2024)
- [M2Diffuser: Diffusion-based Trajectory Optimization for Mobile Manipulation in 3D Scenes](https://mlsanity.com/paper/2410.11402/) (2024)
- [Feat2GS: Probing Visual Foundation Models with Gaussian Splatting](https://mlsanity.com/paper/2412.09606/) (2024)
- [Enhancing Generalization in Vision-Language-Action Models by Preserving Pretrained Representations](https://mlsanity.com/paper/2509.11417/) (2025)
- [Robots Pre-train Robots: Manipulation-Centric Robotic Representation from Large-Scale Robot Datasets](https://mlsanity.com/paper/2410.22325/) (2024)
- [PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation](https://mlsanity.com/paper/2410.01680/) (2024)
- [Parallels Between VLA Model Post-Training and Human Motor Learning: Progress, Challenges, and Trends](https://mlsanity.com/paper/2506.20966/) (2025)
- [In Pursuit of Pixel Supervision for Visual Pre-training](https://mlsanity.com/paper/2512.15715/) (2025)
- [GP3: A 3D Geometry-Aware Policy with Multi-View Images for Robotic Manipulation](https://mlsanity.com/paper/2509.15733/) (2025)
- [On-Robot Reinforcement Learning with Goal-Contrastive Rewards](https://mlsanity.com/paper/2410.19989/) (2024)
- [AnyTask: an Automated Task and Data Generation Framework for Advancing Sim-to-Real Policy Learning](https://mlsanity.com/paper/2512.17853/) (2025)
- [Swiss Army Knife: Synergizing Biases in Knowledge from Vision Foundation Models for Multi-Task Learning](https://mlsanity.com/paper/2410.14633/) (2024)
- [Beyond Cosine Decay: On the effectiveness of Infinite Learning Rate Schedule for Continual Pre-training](https://mlsanity.com/paper/2503.02844/) (2025)
- [FeatSharp: Your Vision Model Features, Sharper](https://mlsanity.com/paper/2502.16025/) (2025)
- [VER: Vision Expert Transformer for Robot Learning via Foundation Distillation and Dynamic Routing](https://mlsanity.com/paper/2510.05213/) (2025)
- [RADSeg: Unleashing Parameter and Compute Efficient Zero-Shot Open-Vocabulary Segmentation Using Agglomerative Models](https://mlsanity.com/paper/2511.19704/) (2025)
- [Mind and Motion Aligned: A Joint Evaluation IsaacSim Benchmark for Task Planning and Low-Level Policies in Mobile Manipulation](https://mlsanity.com/paper/2508.15663/) (2025)
- [M3: 3D-Spatial MultiModal Memory](https://mlsanity.com/paper/2503.16413/) (2025)
- [Order Matters: On Parameter-Efficient Image-to-Video Probing for Recognizing Nearly Symmetric Actions](https://mlsanity.com/paper/2503.24298/) (2025)
- [Slot-Level Robotic Placement via Visual Imitation from Single Human Video](https://mlsanity.com/paper/2504.01959/) (2025)
- [Agglomerating Large Vision Encoders via Distillation for VFSS Segmentation](https://mlsanity.com/paper/2504.02351/) (2025)
- [Is an object-centric representation beneficial for robotic manipulation ?](https://mlsanity.com/paper/2506.19408/) (2025)
- [Model Parallelism With Subnetwork Data Parallelism](https://mlsanity.com/paper/2507.09029/) (2025)
- [SigLino: Efficient Multi-Teacher Distillation for Agglomerative Vision Foundation Models](https://mlsanity.com/paper/2512.20157/) (2025)
- [DeFM: Learning Foundation Representations from Depth for Robotics](https://mlsanity.com/paper/2601.18923/) (2026)

## Abstract (arXiv)

Vision-based robot policy learning, which maps visual inputs to actions, necessitates a holistic understanding of diverse visual tasks beyond single-task needs like classification or segmentation. Inspired by this, we introduce Theia, a vision foundation model for robot learning that distills multiple off-the-shelf vision foundation models trained on varied vision tasks. Theia's rich visual representations encode diverse visual knowledge, enhancing downstream robot learning. Extensive experiments demonstrate that Theia outperforms its teacher models and prior robot learning models using less training data and smaller model sizes. Additionally, we quantify the quality of pre-trained visual representations and hypothesize that higher entropy in feature norm distributions leads to improved robot learning performance. Code, models, and demo are available at https://theia.theaiinstitute.com.