# Robota-Embed: Unified VLA Embeddings for Robot In-Context Learning and Evolution

arXiv: https://arxiv.org/abs/2610.19495 · HTML: https://mlsanity.com/paper/2610.19495/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 35 of its 39 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Builds on (most cited first, 30 of 35)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [ImageBind: One Embedding Space To Bind Them All](https://mlsanity.com/paper/2305.05665/) (2023)
- [SAM 3: Segment Anything with Concepts](https://mlsanity.com/paper/2511.16719/) (2025)
- [DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset](https://mlsanity.com/paper/2403.12945/) (2024)
- [World Action Models are Zero-shot Policies](https://mlsanity.com/paper/2602.15922/) (2026)
- [Fast-WAM: Do World Action Models Need Test-time Future Imagination?](https://mlsanity.com/paper/2603.16666/) (2026)
- [Perception Encoder: The best visual embeddings are not at the output of the network](https://mlsanity.com/paper/2504.13181/) (2025)
- [Causal World Modeling for Robot Control](https://mlsanity.com/paper/2601.21998/) (2026)
- [InternVideo2: Scaling Foundation Models for Multimodal Video Understanding](https://mlsanity.com/paper/2403.15377/) (2024)
- [Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking](https://mlsanity.com/paper/2601.04720/) (2026)
- [ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI](https://mlsanity.com/paper/2410.00425/) (2024)
- [EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video](https://mlsanity.com/paper/2505.11709/) (2025)
- [RoboMIND: Benchmark on Multi-embodiment Intelligence Normative Data for Robot Manipulation](https://mlsanity.com/paper/2412.13877/) (2024)
- [Cosmos 3: Omnimodal World Models for Physical AI](https://mlsanity.com/paper/2606.02800/) (2026)
- [VLM2Vec-V2: Advancing Multimodal Embedding for Videos, Images, and Visual Documents](https://mlsanity.com/paper/2507.04590/) (2025)
- [R+X: Retrieval and Execution from Everyday Human Videos](https://mlsanity.com/paper/2407.12957/) (2024)
- [Retrieval-Augmented Embodied Agents](https://mlsanity.com/paper/2404.11699/) (2024)
- [RICL: Adding In-Context Adaptability to Pre-Trained Vision-Language-Action Models](https://mlsanity.com/paper/2508.02062/) (2025)
- [Contrastive Language, Action, and State Pre-training for Robot Learning](https://mlsanity.com/paper/2304.10782/) (2023)
- [REGENT: A Retrieval-Augmented Generalist Agent That Can Act In-Context in New Environments](https://mlsanity.com/paper/2412.04759/) (2024)
- [Omni-Embed-Nemotron: A Unified Multimodal Retrieval Model for Text, Image, Audio, and Video](https://mlsanity.com/paper/2510.03458/) (2025)
- [T-Rex: Tactile-Reactive Dexterous Manipulation](https://mlsanity.com/paper/2606.17055/) (2026)
- [Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task Generalization](https://mlsanity.com/paper/2608.26103/) (2026)
- [Gemini Embedding 2: A Native Multimodal Embedding Model from Gemini](https://mlsanity.com/paper/2605.27295/) (2026)
- [WeMM-Embedding: WeChat Multi-Modal Embedding Technical Report](https://mlsanity.com/paper/2608.24053/) (2026)
- [HOST:Robots Acquire Manipulation Skills in Seconds from a Single Human Video](https://mlsanity.com/paper/2607.20033/) (2026)
- [VLA-R: Vision-Language Action Retrieval toward Open-World End-to-End Autonomous Driving](https://mlsanity.com/paper/2511.12405/) (2025)

## Abstract (arXiv)

Retrieving useful context for a robot requires matching its task, execution stage, and the continuation that should follow. We introduce Robota-Embed, to our knowledge the first vision-language-action (VLA) embedding model to jointly encode all three modalities in both current-state queries and recorded future continuations. Predictive retrieval training aligns these endpoints, enabling historical continuations to serve as context at inference time. We construct a multi-source corpus of 1,695,684 cleaned training anchors and introduce Robota-Embed-Bench, with 4,000 queries and a shared 8,000-entry gallery. Robota-Embed achieves 47.33% clip-level and 81.77% episode-level R@1, exceeding action-free Gemini Embedding 2 by 22.09 and 33.47 percentage points. Controlled retrieval studies show that removing native action degrades retrieval more than removing vision. We study two complementary applications. Embed2ICL uses retrieved experience for training-free and training-based policy conditioning, increasing success from 47.0% to 56.3% in DROID-Sim and from 86.4% to 92.8% on the LIBERO difficult subset, while improving task progress on real robots. Embed2Evolve uses execution feedback to grow task-local memory and refine context use around a frozen GPT-6 Astra policy. Across five ManiSkill tasks, it achieves 68% success, compared with 24% for GPT-6 Astra Direct and 33% for a harness independently evolved with Omni-Embed-Nemotron-3B under the same protocol. Relative to the latter, it uses approximately 79.5% fewer policy tokens per test rollout, counting both successes and failures. These results show that a unified, action-aware retrieval space can support robot in-context learning and task-local context evolution.