---
title: "FAST: Efficient Action Tokenization for Vision-Language-Action Models — Pertsch 2025"
description: "Instead, the authors propose FAST. 75 references, cited by 38 papers here."
authors: ["Karl Pertsch", "Kyle Stachowicz", "Brian Ichter", "Danny Driess", "Suraj Nair", "Quan Vuong", "Oier Mees", "Chelsea Finn", "Sergey Levine"]
year: 2025
venue: "arXiv:2501.09747"
source: "https://arxiv.org/abs/2501.09747"
doi: "10.48550/arxiv.2501.09747"
arxiv: "2501.09747"
licence: "CC BY 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/pertsch-2025-fast-efficient-action-tokenization-vision/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/pertsch-2025-fast-efficient-action-tokenization-vision/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/) / FAST: Efficient Action Tokenization for Vision-Language-Action Models

# FAST: Efficient Action Tokenization for Vision-Language-Action Models

**Authors:** Karl Pertsch, Kyle Stachowicz, [Brian Ichter](https://md.research.elamur.ai/free/en/authors/brian-ichter/), Danny Driess, [Suraj Nair](https://md.research.elamur.ai/free/en/authors/suraj-nair/), Quan Vuong, Oier Mees, [Chelsea Finn](https://md.research.elamur.ai/free/en/authors/chelsea-finn/) and [Sergey Levine](https://md.research.elamur.ai/free/en/authors/sergey-levine/) · **Year:** 2025 · **Venue:** arXiv:2501.09747 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2501.09747) · [PDF](https://arxiv.org/pdf/2501.09747) · [DOI 10.48550/arxiv.2501.09747](https://doi.org/10.48550/arxiv.2501.09747) · [arXiv 2501.09747](https://arxiv.org/abs/2501.09747)

**Citations:** 38 in our library, of the 140,884 papers whose bibliographies we read

## Summary
The authors investigate how to convert continuous robot control signals into discrete tokens for autoregressive VLA models. The common approach quantizes each action dimension separately at every time step, usually into 256 bins. On high-frequency data it performs poorly: neighboring tokens barely differ, so the model learns to simply repeat the previous token and gets stuck in a poor local optimum. Instead, the authors propose FAST. Before training, an entire segment of future actions (a "chunk") is compressed with the discrete cosine transform (DCT), as JPEG does with images. The idea is similar to text compression via BPE in language models. This reduces both the correlation between tokens and their number. Building on the method, the authors released FAST+, a universal tokenizer trained on one million real trajectories from various robots. Paired with π0 (the π0-FAST model), the approach scales to 10 thousand hours of data. It matches the diffusion-based π0 in performance while training up to 5 times faster. Thus, standard next-token prediction is now suitable for dexterous, high-frequency manipulation.

## Key points
- The baseline action tokenization approach, as in OpenVLA, produces hundreds of tokens per action chunk on high-frequency data. On a synthetic cubic spline interpolation task, its performance degrades as the sampling rate increases, whereas FAST remains stable.
- FAST+ is trained on 1 million trajectories and can be used as an off-the-shelf tokenizer for very different robots. These include single-arm manipulators (Franka, UR5, WidowX), the bimanual ARX and ViperX (ALOHA) systems, ALLEGRO hands, the Unitree H1 humanoid, UMI, and even Waymo self-driving car data.
- π0-FAST is trained on all successful DROID episodes: 75 thousand episodes, 21 million samples, 240 thousand iterations. According to the authors, it is the first generalist language-conditioned policy that works zero-shot in new environments. The quantitative evaluation covered 16 tasks with 44 trials, and qualitative tests were conducted on the campuses of three universities.
- The robots were evaluated on the following tasks: table clearing (Table Bussing), T-shirt folding (T-Shirt Folding), grocery packing (Grocery Bagging), removing toast from a toaster, and laundry folding (Laundry Folding).
- Under the same compute budget, π0-FAST clearly outperforms the diffusion-based π0 because it converges faster. FAST requires no changes to the architecture of the pretrained transformer.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [AI & machine learning](https://md.research.elamur.ai/free/en/shelves/ai/)

Topics: [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · [✂️ Tokenization & segmentation](https://md.research.elamur.ai/free/en/topics/tokenization/)

## Cite

Cite as: Karl Pertsch, Kyle Stachowicz, Brian Ichter, Danny Driess, Suraj Nair, Quan Vuong, Oier Mees, Chelsea Finn and Sergey Levine. 2025. FAST: Efficient Action Tokenization for Vision-Language-Action Models. arXiv:2501.09747.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@misc{pertsch2025fast,
  title = {{FAST: Efficient Action Tokenization for Vision-Language-Action Models}},
  author = {Karl Pertsch, Kyle Stachowicz, Brian Ichter, Danny Driess, Suraj Nair, Quan Vuong, Oier Mees, Chelsea Finn and Sergey Levine},
  year = {2025},
  doi = {10.48550/arxiv.2501.09747},
  eprint = {2501.09747},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2501.09747}
}
```

- [← previous Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling](https://md.research.elamur.ai/free/en/papers/liu-2025-1b-llm-surpass-405b-llm/)
- [next → OG-RAG: Ontology-grounded retrieval-augmented generation for large language models](https://md.research.elamur.ai/free/en/papers/sharma-2025-og-rag-ontology-grounded-retrieval/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/)

## References

75 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Nasir Ahmed, T Natarajan, and Kamisetty R Rao. Discrete cosine transform. IEEE transactions on Computers, 100(1):90–93, 1974. [find](https://scholar.google.com/scholar?q=Nasir+Ahmed%2C+T+Natarajan%2C+and+Kamisetty+R+Rao%2E+Discrete+cosine+transform%2E+IEEE+transactions+on+Computers%2C+100%281%29%3A90%E2%80%9393%2C+1974%2E)
2. Michael Ahn, Anthony Brohan, Noah Brown, Yevgen Chebotar, Omar Cortes, Byron David, Chelsea Finn, Chuyuan Fu, Keerthana Gopalakrishnan, Karol Hausman, Alex Herzog, Daniel Ho, Jasmine Hsu, Julian Ibarz, Brian Ichter, Alex Irpan, Eric Jang, Rosario Jauregui Ruano, Kyle Jeffrey, Sally Jesmonth, Nikhil… [our card](https://md.research.elamur.ai/free/en/papers/ahn-2022-i-i-say-grounding-language/)
3. Suneel Belkhale and Dorsa Sadigh. Minivla: A better vla with a smaller footprint, 2024. URL [https://github.com/](https://github.com/) Stanford-ILIAD/openvla-mini. [↗](https://github.com/)
4. Suneel Belkhale, Tianli Ding, Ted Xiao, Pierre Sermanet, Quon Vuong, Jonathan Tompson, Yevgen Chebotar, De-bidatta Dwibedi, and Dorsa Sadigh. Rt-h: Action hierarchies using language, 2024. URL https://arxiv.org/abs/ 2403.01823. [↗](https://arxiv.org/abs/)
5. Lucas Beyer, Andreas Steiner, André Susano Pinto, Alexander Kolesnikov, Xiao Wang, Daniel Salz, Maxim Neumann, Ibrahim Alabdulmohsin, Michael Tschannen, Emanuele Bugliarello, et al. Paligemma: A versatile 3b vlm for transfer. arXiv preprint arXiv:2407.07726, 2024. [find](https://scholar.google.com/scholar?q=Lucas+Beyer%2C+Andreas+Steiner%2C+Andre%CC%81+Susano+Pinto%2C+Alexander+Kolesnikov%2C+Xiao+Wang%2C+Daniel+Salz%2C+Maxim+Neumann%2C+Ibrahim+Alabdulmohsin%2C+Michael+Tschannen%2C+Emanuele+Bugliarello%2C+et+al%2E+Paligemma%3A+A+vers)
6. Homanga Bharadhwaj, Jay Vakil, Mohit Sharma, Abhinav Gupta, Shubham Tulsiani, and Vikash Kumar. Roboagent: Generalization and efficiency in robot manipulation via semantic augmentations and action chunking. In 2024 IEEE International Conference on Robotics and Automation (ICRA), pages 4788–4795. IEE… [find](https://scholar.google.com/scholar?q=Homanga+Bharadhwaj%2C+Jay+Vakil%2C+Mohit+Sharma%2C+Abhinav+Gupta%2C+Shubham+Tulsiani%2C+and+Vikash+Kumar%2E+Roboagent%3A+Generalization+and+efficiency+in+robot+manipulation+via+semantic+augmentations+and+action+chu)
7. Kevin Black, Noah Brown, Danny Driess, Adnan Esmail, Michael Equi, Chelsea Finn, Niccolo Fusai, Lachy Groom, Karol Hausman, Brian Ichter, et al. pi 0: A vision-language-action flow model for general robot control. arXiv preprint arXiv:2410.24164, 2024. [find](https://scholar.google.com/scholar?q=Kevin+Black%2C+Noah+Brown%2C+Danny+Driess%2C+Adnan+Esmail%2C+Michael+Equi%2C+Chelsea+Finn%2C+Niccolo+Fusai%2C+Lachy+Groom%2C+Karol+Hausman%2C+Brian+Ichter%2C+et+al%2E+pi+0%3A+A+vision%2Dlanguage%2Daction+flow+model+for+general+r)
8. Anthony Brohan, Noah Brown, Justice Carbajal, Yevgen Chebotar, Joseph Dabis, Chelsea Finn, Keerthana Gopalakrishnan, Karol Hausman, Alex Herzog, Jasmine Hsu, Julian Ibarz, Brian Ichter, Alex Irpan, Tomas Jackson, Sally Jesmonth, Nikhil Joshi, Ryan Julian, Dmitry Kalashnikov, Yuheng Kuang, Isabel Lea… [find](https://scholar.google.com/scholar?q=Anthony+Brohan%2C+Noah+Brown%2C+Justice+Carbajal%2C+Yevgen+Chebotar%2C+Joseph+Dabis%2C+Chelsea+Finn%2C+Keerthana+Gopalakrishnan%2C+Karol+Hausman%2C+Alex+Herzog%2C+Jasmine+Hsu%2C+Julian+Ibarz%2C+Brian+Ichter%2C+Alex+Irpan%2C+To)
9. Anthony Brohan, Noah Brown, Justice Carbajal, Yevgen Chebotar, Joseph Dabis, Chelsea Finn, Keerthana Gopalakrishnan, Karol Hausman, Alex Herzog, Jasmine Hsu, et al. Rt-1: Robotics transformer for real-world control at scale. arXiv preprint arXiv:2212.06817, 2022. [find](https://scholar.google.com/scholar?q=Anthony+Brohan%2C+Noah+Brown%2C+Justice+Carbajal%2C+Yevgen+Chebotar%2C+Joseph+Dabis%2C+Chelsea+Finn%2C+Keerthana+Gopalakrishnan%2C+Karol+Hausman%2C+Alex+Herzog%2C+Jasmine+Hsu%2C+et+al%2E+Rt%2D1%3A+Robotics+transformer+for+real)
10. Anthony Brohan, Noah Brown, Justice Carbajal, Yevgen Chebotar, Xi Chen, Krzysztof Choromanski, Tianli Ding, Danny Driess, Avinava Dubey, Chelsea Finn, Pete Florence, Chuyuan Fu, Montse Gonzalez Arenas, Keerthana Gopalakrishnan, Kehang Han, Karol Hausman, Alex Herzog, Jasmine Hsu, Brian Ichter, Alex… [find](https://scholar.google.com/scholar?q=Anthony+Brohan%2C+Noah+Brown%2C+Justice+Carbajal%2C+Yevgen+Chebotar%2C+Xi+Chen%2C+Krzysztof+Choromanski%2C+Tianli+Ding%2C+Danny+Driess%2C+Avinava+Dubey%2C+Chelsea+Finn%2C+Pete+Florence%2C+Chuyuan+Fu%2C+Montse+Gonzalez+Arenas)
11. Chi-Lam Cheang, Guangzeng Chen, Ya Jing, Tao Kong, Hang Li, Yifeng Li, Yuxiao Liu, Hongtao Wu, Jiafeng Xu, Yichu Yang, Hanbo Zhang, and Minzhao Zhu. Gr-2: A generative video-language-action model with web-scale knowledge for robot manipulation. arXiv preprint arXiv:2410.06158, 2024. — [arXiv:2410.06158](https://arxiv.org/abs/2410.06158) [find](https://scholar.google.com/scholar?q=Chi%2DLam+Cheang%2C+Guangzeng+Chen%2C+Ya+Jing%2C+Tao+Kong%2C+Hang+Li%2C+Yifeng+Li%2C+Yuxiao+Liu%2C+Hongtao+Wu%2C+Jiafeng+Xu%2C+Yichu+Yang%2C+Hanbo+Zhang%2C+and+Minzhao+Zhu%2E+Gr%2D2%3A+A+generative+video%2Dlanguage%2Daction+model+with)
12. Sanyuan Chen, Yu Wu, Chengyi Wang, Shujie Liu, Daniel Tompkins, Zhuo Chen, and Furu Wei. Beats: Audio pre-training with acoustic tokenizers. arXiv preprint arXiv:2212.09058, 2022. — [arXiv:2212.09058](https://arxiv.org/abs/2212.09058) [find](https://scholar.google.com/scholar?q=Sanyuan+Chen%2C+Yu+Wu%2C+Chengyi+Wang%2C+Shujie+Liu%2C+Daniel+Tompkins%2C+Zhuo+Chen%2C+and+Furu+Wei%2E+Beats%3A+Audio+pre%2Dtraining+with+acoustic+tokenizers%2E+arXiv+preprint+arXiv%3A2212%2E09058%2C+2022%2E+%E2%80%94+%5BarXiv%3A2212%2E09058%5D)
13. An-Chieh Cheng, Yandong Ji, Zhaojing Yang, Xueyan Zou, Jan Kautz, Erdem Biyik, Hongxu Yin, Sifei Liu, and Xiaolong Wang. NaVILA: Legged Robot Vision-Language-Action Model for Navigation. arXiv preprint arXiv:2412.04453, 2024. — [arXiv:2412.04453](https://arxiv.org/abs/2412.04453) [find](https://scholar.google.com/scholar?q=An%2DChieh+Cheng%2C+Yandong+Ji%2C+Zhaojing+Yang%2C+Xueyan+Zou%2C+Jan+Kautz%2C+Erdem+Biyik%2C+Hongxu+Yin%2C+Sifei+Liu%2C+and+Xiaolong+Wang%2E+NaVILA%3A+Legged+Robot+Vision%2DLanguage%2DAction+Model+for+Navigation%2E+arXiv+preprin)
14. Xuxin Cheng, Jialong Li, Shiqi Yang, Ge Yang, and Xiaolong Wang. Open-television: Teleoperation with immersive active visual feedback. arXiv preprint arXiv:2407.01512, 2024. — [arXiv:2407.01512](https://arxiv.org/abs/2407.01512) [find](https://scholar.google.com/scholar?q=Xuxin+Cheng%2C+Jialong+Li%2C+Shiqi+Yang%2C+Ge+Yang%2C+and+Xiaolong+Wang%2E+Open%2Dtelevision%3A+Teleoperation+with+immersive+active+visual+feedback%2E+arXiv+preprint+arXiv%3A2407%2E01512%2C+2024%2E+%E2%80%94+%5BarXiv%3A2407%2E01512%5D%28https)
15. Cheng Chi, Siyuan Feng, Yilun Du, Zhenjia Xu, Eric Cousineau, Benjamin Burchfiel, and Shuran Song. Diffusion policy: Visuomotor policy learning via action diffusion. In Proceedings of Robotics: Science and Systems (RSS), 2023. [find](https://scholar.google.com/scholar?q=Cheng+Chi%2C+Siyuan+Feng%2C+Yilun+Du%2C+Zhenjia+Xu%2C+Eric+Cousineau%2C+Benjamin+Burchfiel%2C+and+Shuran+Song%2E+Diffusion+policy%3A+Visuomotor+policy+learning+via+action+diffusion%2E+In+Proceedings+of+Robotics%3A+Scienc)
16. Cheng Chi, Zhenjia Xu, Chuer Pan, Eric Cousineau, Benjamin Burchfiel, Siyuan Feng, Russ Tedrake, and Shuran Song. Universal manipulation interface: In-the-wild robot teaching without in-the-wild robots. In Proceedings of Robotics: Science and Systems (RSS), 2024. [find](https://scholar.google.com/scholar?q=Cheng+Chi%2C+Zhenjia+Xu%2C+Chuer+Pan%2C+Eric+Cousineau%2C+Benjamin+Burchfiel%2C+Siyuan+Feng%2C+Russ+Tedrake%2C+and+Shuran+Song%2E+Universal+manipulation+interface%3A+In%2Dthe%2Dwild+robot+teaching+without+in%2Dthe%2Dwild+robot)
17. OX-Embodiment Collaboration, A Padalkar, A Pooley, A Jain, A Bewley, A Herzog, A Irpan, A Khazatsky, A Rai, A Singh, et al. Open X-Embodiment: Robotic learning datasets and RT-X models. arXiv preprint arXiv:2310.08864, 1(2), 2023. [find](https://scholar.google.com/scholar?q=OX%2DEmbodiment+Collaboration%2C+A+Padalkar%2C+A+Pooley%2C+A+Jain%2C+A+Bewley%2C+A+Herzog%2C+A+Irpan%2C+A+Khazatsky%2C+A+Rai%2C+A+Singh%2C+et+al%2E+Open+X%2DEmbodiment%3A+Robotic+learning+datasets+and+RT%2DX+models%2E+arXiv+preprint)
18. James W Cooley and John W Tukey. An algorithm for the machine calculation of complex fourier series. Mathematics of computation, 19(90):297–301, 1965. [find](https://scholar.google.com/scholar?q=James+W+Cooley+and+John+W+Tukey%2E+An+algorithm+for+the+machine+calculation+of+complex+fourier+series%2E+Mathematics+of+computation%2C+19%2890%29%3A297%E2%80%93301%2C+1965%2E)
19. Norman Di Palo and Edward Johns. Keypoint action tokens enable in-context imitation learning in robotics. In Proceedings of Robotics: Science and Systems (RSS), 2024. [find](https://scholar.google.com/scholar?q=Norman+Di+Palo+and+Edward+Johns%2E+Keypoint+action+tokens+enable+in%2Dcontext+imitation+learning+in+robotics%2E+In+Proceedings+of+Robotics%3A+Science+and+Systems+%28RSS%29%2C+2024%2E)
20. Ria Doshi, Homer Walke, Oier Mees, Sudeep Dasari, and Sergey Levine. Scaling cross-embodied learning: One policy for manipulation, navigation, locomotion and aviation. In Conference on Robot Learning, 2024. [find](https://scholar.google.com/scholar?q=Ria+Doshi%2C+Homer+Walke%2C+Oier+Mees%2C+Sudeep+Dasari%2C+and+Sergey+Levine%2E+Scaling+cross%2Dembodied+learning%3A+One+policy+for+manipulation%2C+navigation%2C+locomotion+and+aviation%2E+In+Conference+on+Robot+Learning%2C)
21. Danny Driess, Fei Xia, Mehdi SM Sajjadi, Corey Lynch, Aakanksha Chowdhery, Brian Ichter, Ayzaan Wahid, Jonathan Tompson, Quan Vuong, Tianhe Yu, et al. Palm-e: An embodied multimodal language model. arXiv preprint arXiv:2303.03378, 2023. [find](https://scholar.google.com/scholar?q=Danny+Driess%2C+Fei+Xia%2C+Mehdi+SM+Sajjadi%2C+Corey+Lynch%2C+Aakanksha+Chowdhery%2C+Brian+Ichter%2C+Ayzaan+Wahid%2C+Jonathan+Tompson%2C+Quan+Vuong%2C+Tianhe+Yu%2C+et+al%2E+Palm%2De%3A+An+embodied+multimodal+language+model%2E+ar)
22. Patrick Esser, Robin Rombach, and Björn Ommer. Taming transformers for high-resolution image synthesis, 2020. [find](https://scholar.google.com/scholar?q=Patrick+Esser%2C+Robin+Rombach%2C+and+Bjo%CC%88rn+Ommer%2E+Taming+transformers+for+high%2Dresolution+image+synthesis%2C+2020%2E)
23. Scott Ettinger, Shuyang Cheng, Benjamin Caine, Chenxi Liu, Hang Zhao, Sabeek Pradhan, Yuning Chai, Ben Sapp, Charles R. Qi, Yin Zhou, Zoey Yang, Aur’elien Chouard, Pei Sun, Jiquan Ngiam, Vijay Vasudevan, Alexander McCauley, Jonathon Shlens, and Dragomir Anguelov. Large scale interactive motion forec… [find](https://scholar.google.com/scholar?q=Scott+Ettinger%2C+Shuyang+Cheng%2C+Benjamin+Caine%2C+Chenxi+Liu%2C+Hang+Zhao%2C+Sabeek+Pradhan%2C+Yuning+Chai%2C+Ben+Sapp%2C+Charles+R%2E+Qi%2C+Yin+Zhou%2C+Zoey+Yang%2C+Aur%E2%80%99elien+Chouard%2C+Pei+Sun%2C+Jiquan+Ngiam%2C+Vijay+Vasudev)
24. Hao-Shu Fang, Hongjie Fang, Zhenyu Tang, Jirong Liu, Chenxi Wang, Junbo Wang, Haoyi Zhu, and Cewu Lu. Rh20t: A comprehensive robotic dataset for learning diverse skills in one-shot. In 2024 IEEE International Conference on Robotics and Automation (ICRA), pages 653–660. IEEE, 2024. [find](https://scholar.google.com/scholar?q=Hao%2DShu+Fang%2C+Hongjie+Fang%2C+Zhenyu+Tang%2C+Jirong+Liu%2C+Chenxi+Wang%2C+Junbo+Wang%2C+Haoyi+Zhu%2C+and+Cewu+Lu%2E+Rh20t%3A+A+comprehensive+robotic+dataset+for+learning+diverse+skills+in+one%2Dshot%2E+In+2024+IEEE+Inter)
25. Kuan Fang, Fangchen Liu, Pieter Abbeel, and Sergey Levine. Moka: Open-world robotic manipulation through mark-based visual prompting. Robotics: Science and Systems (RSS), 2024. [find](https://scholar.google.com/scholar?q=Kuan+Fang%2C+Fangchen+Liu%2C+Pieter+Abbeel%2C+and+Sergey+Levine%2E+Moka%3A+Open%2Dworld+robotic+manipulation+through+mark%2Dbased+visual+prompting%2E+Robotics%3A+Science+and+Systems+%28RSS%29%2C+2024%2E)
26. Zipeng Fu, Qingqing Zhao, Qi Wu, Gordon Wetzstein, and Chelsea Finn. Humanplus: Humanoid shadowing and imitation from humans. In Conference on Robot Learning (CoRL), 2024. [find](https://scholar.google.com/scholar?q=Zipeng+Fu%2C+Qingqing+Zhao%2C+Qi+Wu%2C+Gordon+Wetzstein%2C+and+Chelsea+Finn%2E+Humanplus%3A+Humanoid+shadowing+and+imitation+from+humans%2E+In+Conference+on+Robot+Learning+%28CoRL%29%2C+2024%2E)
27. Philip Gage. A new algorithm for data compression. The C Users Journal, 12(2):23–38, 1994. [find](https://scholar.google.com/scholar?q=Philip+Gage%2E+A+new+algorithm+for+data+compression%2E+The+C+Users+Journal%2C+12%282%29%3A23%E2%80%9338%2C+1994%2E)
28. Dan Gillick, Cliff Brunk, Oriol Vinyals, and Amarnag Subramanya. Multilingual language processing from bytes, 2016. URL https://arxiv.org/abs/1512.00103. [↗](https://arxiv.org/abs/1512.00103)
29. Yuan Gong, Yu-An Chung, and James Glass. AST: Audio Spectrogram Transformer. In Proc. Interspeech 2021, pages 571–575, 2021. doi: 10.21437/Interspeech. 2021-698. — [doi:10.21437/interspeech](https://doi.org/10.21437/interspeech) [find](https://scholar.google.com/scholar?q=Yuan+Gong%2C+Yu%2DAn+Chung%2C+and+James+Glass%2E+AST%3A+Audio+Spectrogram+Transformer%2E+In+Proc%2E+Interspeech+2021%2C+pages+571%E2%80%93575%2C+2021%2E+doi%3A+10%2E21437%2FInterspeech%2E+2021%2D698%2E+%E2%80%94+%5Bdoi%3A10%2E21437%2Finterspeech%5D%28https%3A%2F%2Fd)
30. Irmak Guzey, Yinlong Dai, Georgy Savva, Raunaq Bhi-rangi, and Lerrel Pinto. Bridging the human to robot dexterity gap through object-oriented rewards, 2024. URL https://arxiv.org/abs/2410.23289. — [arXiv:2410.23289](https://arxiv.org/abs/2410.23289) [↗](https://arxiv.org/abs/2410.23289)
31. Huy Ha, Yihuai Gao, Zipeng Fu, Jie Tan, and Shuran Song. UMI on legs: Making manipulation policies mobile with manipulation-centric whole-body controllers. In Proceedings of the 2024 Conference on Robot Learning, 2024. [find](https://scholar.google.com/scholar?q=Huy+Ha%2C+Yihuai+Gao%2C+Zipeng+Fu%2C+Jie+Tan%2C+and+Shuran+Song%2E+UMI+on+legs%3A+Making+manipulation+policies+mobile+with+manipulation%2Dcentric+whole%2Dbody+controllers%2E+In+Proceedings+of+the+2024+Conference+on+Rob)
32. Wenlong Huang, Chen Wang, Yunzhu Li, Ruohan Zhang, and Li Fei-Fei. Rekep: Spatio-temporal reasoning of relational keypoint constraints for robotic manipulation. arXiv preprint arXiv:2409.01652, 2024. — [arXiv:2409.01652](https://arxiv.org/abs/2409.01652) [find](https://scholar.google.com/scholar?q=Wenlong+Huang%2C+Chen+Wang%2C+Yunzhu+Li%2C+Ruohan+Zhang%2C+and+Li+Fei%2DFei%2E+Rekep%3A+Spatio%2Dtemporal+reasoning+of+relational+keypoint+constraints+for+robotic+manipulation%2E+arXiv+preprint+arXiv%3A2409%2E01652%2C+2024%2E+)
33. David A. Huffman. A method for the construction of minimum-redundancy codes. Proceedings of the IRE, 40(9):1098–1101, 1952. doi: 10.1109/JRPROC.1952. 273898. [find](https://scholar.google.com/scholar?q=David+A%2E+Huffman%2E+A+method+for+the+construction+of+minimum%2Dredundancy+codes%2E+Proceedings+of+the+IRE%2C+40%289%29%3A1098%E2%80%931101%2C+1952%2E+doi%3A+10%2E1109%2FJRPROC%2E1952%2E+273898%2E)
34. Huiwon Jang, Sihyun Yu, Jinwoo Shin, Pieter Abbeel, and Younggyo Seo. Efficient long video tokenization via coordinated-based patch reconstruction. arXiv preprint arXiv:2411.14762, 2024. — [arXiv:2411.14762](https://arxiv.org/abs/2411.14762) [find](https://scholar.google.com/scholar?q=Huiwon+Jang%2C+Sihyun+Yu%2C+Jinwoo+Shin%2C+Pieter+Abbeel%2C+and+Younggyo+Seo%2E+Efficient+long+video+tokenization+via+coordinated%2Dbased+patch+reconstruction%2E+arXiv+preprint+arXiv%3A2411%2E14762%2C+2024%2E+%E2%80%94+%5BarXiv%3A2411)
35. Zhenyu Jiang, Yuqi Xie, Kevin Lin, Zhenjia Xu, Weikang Wan, Ajay Mandlekar, Linxi Fan, and Yuke Zhu. Dexmimicgen: Automated data generation for bimanual dexterous manipulation via imitation learning. arXiv preprint arXiv:2410.24185, 2024. — [arXiv:2410.24185](https://arxiv.org/abs/2410.24185) [find](https://scholar.google.com/scholar?q=Zhenyu+Jiang%2C+Yuqi+Xie%2C+Kevin+Lin%2C+Zhenjia+Xu%2C+Weikang+Wan%2C+Ajay+Mandlekar%2C+Linxi+Fan%2C+and+Yuke+Zhu%2E+Dexmimicgen%3A+Automated+data+generation+for+bimanual+dexterous+manipulation+via+imitation+learning%2E+)
36. Joshua Jones, Oier Mees, Carmelo Sferrazza, Kyle Stachowicz, Pieter Abbeel, and Sergey Levine. Beyond sight: Finetuning generalist robot policies with heterogeneous sensors via language grounding. arXiv preprint arXiv:2501.04693, 2025. — [arXiv:2501.04693](https://arxiv.org/abs/2501.04693) [find](https://scholar.google.com/scholar?q=Joshua+Jones%2C+Oier+Mees%2C+Carmelo+Sferrazza%2C+Kyle+Stachowicz%2C+Pieter+Abbeel%2C+and+Sergey+Levine%2E+Beyond+sight%3A+Finetuning+generalist+robot+policies+with+heterogeneous+sensors+via+language+grounding%2E+arX)
37. Siddharth Karamcheti, Suraj Nair, Ashwin Balakrishna, Percy Liang, Thomas Kollar, and Dorsa Sadigh. Prismatic vlms: Investigating the design space of visually-conditioned language models. In International Conference on Machine Learning (ICML), 2024. [find](https://scholar.google.com/scholar?q=Siddharth+Karamcheti%2C+Suraj+Nair%2C+Ashwin+Balakrishna%2C+Percy+Liang%2C+Thomas+Kollar%2C+and+Dorsa+Sadigh%2E+Prismatic+vlms%3A+Investigating+the+design+space+of+visually%2Dconditioned+language+models%2E+In+Internati)
38. Alexander Khazatsky, Karl Pertsch, Suraj Nair, Ashwin Balakrishna, Sudeep Dasari, Siddharth Karamcheti, Soroush Nasiriany, Mohan Kumar Srirama, Lawrence Yunliang Chen, Kirsty Ellis, Peter David Fagan, Joey Hejna, Masha Itkina, Marion Lepert, Yecheng Jason Ma, Patrick Tree Miller, Jimmy Wu, Suneel Be… [find](https://scholar.google.com/scholar?q=Alexander+Khazatsky%2C+Karl+Pertsch%2C+Suraj+Nair%2C+Ashwin+Balakrishna%2C+Sudeep+Dasari%2C+Siddharth+Karamcheti%2C+Soroush+Nasiriany%2C+Mohan+Kumar+Srirama%2C+Lawrence+Yunliang+Chen%2C+Kirsty+Ellis%2C+Peter+David+Fagan%2C)
39. Moo Jin Kim, Karl Pertsch, Siddharth Karamcheti, Ted Xiao, Ashwin Balakrishna, Suraj Nair, Rafael Rafailov, Ethan Foster, Grace Lam, Pannag Sanketi, et al. Openvla: An open-source vision-language-action model. arXiv preprint arXiv:2406.09246, 2024. [find](https://scholar.google.com/scholar?q=Moo+Jin+Kim%2C+Karl+Pertsch%2C+Siddharth+Karamcheti%2C+Ted+Xiao%2C+Ashwin+Balakrishna%2C+Suraj+Nair%2C+Rafael+Rafailov%2C+Ethan+Foster%2C+Grace+Lam%2C+Pannag+Sanketi%2C+et+al%2E+Openvla%3A+An+open%2Dsource+vision%2Dlanguage%2Dacti)
40. Lucy Lai, Ann ZX Huang, and Samuel J Gershman. Action chunking as conditional policy compression. [find](https://scholar.google.com/scholar?q=Lucy+Lai%2C+Ann+ZX+Huang%2C+and+Samuel+J+Gershman%2E+Action+chunking+as+conditional+policy+compression%2E)
41. Seungjae Lee, Yibin Wang, Haritheja Etukuru, H. Jin Kim, Nur Muhammad Mahi Shafiullah, and Lerrel Pinto. Behavior generation with latent actions. arXiv preprint arXiv:2403.03181, 2024. — [arXiv:2403.03181](https://arxiv.org/abs/2403.03181) [find](https://scholar.google.com/scholar?q=Seungjae+Lee%2C+Yibin+Wang%2C+Haritheja+Etukuru%2C+H%2E+Jin+Kim%2C+Nur+Muhammad+Mahi+Shafiullah%2C+and+Lerrel+Pinto%2E+Behavior+generation+with+latent+actions%2E+arXiv+preprint+arXiv%3A2403%2E03181%2C+2024%2E+%E2%80%94+%5BarXiv%3A2403%2E0)
42. Toru Lin, Yu Zhang, Qiyang Li, Haozhi Qi, Brent Yi, Sergey Levine, and Jitendra Malik. Learning visuotactile skills with two multifingered hands. arXiv:2404.16823, 2024. — [arXiv:2404.16823](https://arxiv.org/abs/2404.16823) [find](https://scholar.google.com/scholar?q=Toru+Lin%2C+Yu+Zhang%2C+Qiyang+Li%2C+Haozhi+Qi%2C+Brent+Yi%2C+Sergey+Levine%2C+and+Jitendra+Malik%2E+Learning+visuotactile+skills+with+two+multifingered+hands%2E+arXiv%3A2404%2E16823%2C+2024%2E+%E2%80%94+%5BarXiv%3A2404%2E16823%5D%28https%3A%2F%2Fa)
43. Bo Liu, Yifeng Zhu, Chongkai Gao, Yihao Feng, Qiang Liu, Yuke Zhu, and Peter Stone. Libero: Benchmarking knowledge transfer for lifelong robot learning. Advances in Neural Information Processing Systems, 36, 2024. [find](https://scholar.google.com/scholar?q=Bo+Liu%2C+Yifeng+Zhu%2C+Chongkai+Gao%2C+Yihao+Feng%2C+Qiang+Liu%2C+Yuke+Zhu%2C+and+Peter+Stone%2E+Libero%3A+Benchmarking+knowledge+transfer+for+lifelong+robot+learning%2E+Advances+in+Neural+Information+Processing+Syste)
44. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual instruction tuning. In Advances in Neural Information Processing Systems (NeurIPS), 2023. [our card](https://md.research.elamur.ai/free/en/papers/liu-2023-visual-instruction-tuning/)
45. Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017. [find](https://scholar.google.com/scholar?q=Ilya+Loshchilov+and+Frank+Hutter%2E+Decoupled+weight+decay+regularization%2E+arXiv+preprint+arXiv%3A1711%2E05101%2C+2017%2E)
46. Jianlan Luo, Zheyuan Hu, Charles Xu, You Liang Tan, Jacob Berg, Archit Sharma, Stefan Schaal, Chelsea Finn, Abhishek Gupta, and Sergey Levine. Serl: A software suite for sample-efficient robotic reinforcement learning, 2024. [find](https://scholar.google.com/scholar?q=Jianlan+Luo%2C+Zheyuan+Hu%2C+Charles+Xu%2C+You+Liang+Tan%2C+Jacob+Berg%2C+Archit+Sharma%2C+Stefan+Schaal%2C+Chelsea+Finn%2C+Abhishek+Gupta%2C+and+Sergey+Levine%2E+Serl%3A+A+software+suite+for+sample%2Defficient+robotic+reinf)
47. Ajay Mandlekar, Yuke Zhu, Animesh Garg, Jonathan Booher, Max Spero, Albert Tung, Julian Gao, John Emmons, Anchit Gupta, Emre Orbay, et al. Roboturk: A crowdsourcing platform for robotic skill learning through imitation. In Conference on Robot Learning, pages 879– 893. PMLR, 2018. [find](https://scholar.google.com/scholar?q=Ajay+Mandlekar%2C+Yuke+Zhu%2C+Animesh+Garg%2C+Jonathan+Booher%2C+Max+Spero%2C+Albert+Tung%2C+Julian+Gao%2C+John+Emmons%2C+Anchit+Gupta%2C+Emre+Orbay%2C+et+al%2E+Roboturk%3A+A+crowdsourcing+platform+for+robotic+skill+learning)
48. Fabian Mentzer, David Minnen, Eirikur Agustsson, and Michael Tschannen. Finite scalar quantization: Vq-vae made simple, 2023. URL https://arxiv.org/abs/2309. 15505. [↗](https://arxiv.org/abs/2309)
49. Atharva Mete, Haotian Xue, Albert Wilcox, Yongxin Chen, and Animesh Garg. Quest: Self-supervised skill abstractions for learning continuous control, 2024. URL https://arxiv.org/abs/2407.15840. — [arXiv:2407.15840](https://arxiv.org/abs/2407.15840) [↗](https://arxiv.org/abs/2407.15840)
50. Soroush Nasiriany, Fei Xia, Wenhao Yu, Ted Xiao, Jacky Liang, Ishita Dasgupta, Annie Xie, Danny Driess, Ayzaan Wahid, Zhuo Xu, et al. Pivot: Iterative visual prompting elicits actionable knowledge for vlms. In Forty-first International Conference on Machine Learning, 2024. [find](https://scholar.google.com/scholar?q=Soroush+Nasiriany%2C+Fei+Xia%2C+Wenhao+Yu%2C+Ted+Xiao%2C+Jacky+Liang%2C+Ishita+Dasgupta%2C+Annie+Xie%2C+Danny+Driess%2C+Ayzaan+Wahid%2C+Zhuo+Xu%2C+et+al%2E+Pivot%3A+Iterative+visual+prompting+elicits+actionable+knowledge+for)
51. Octo Model Team, Dibya Ghosh, Homer Walke, Karl Pertsch, Kevin Black, Oier Mees, Sudeep Dasari, Joey Hejna, Charles Xu, Jianlan Luo, Tobias Kreiman, You Liang Tan, Pannag Sanketi, Quan Vuong, Ted Xiao, Dorsa Sadigh, Chelsea Finn, and Sergey Levine. Octo: An open-source generalist robot policy. In Pr… [find](https://scholar.google.com/scholar?q=Octo+Model+Team%2C+Dibya+Ghosh%2C+Homer+Walke%2C+Karl+Pertsch%2C+Kevin+Black%2C+Oier+Mees%2C+Sudeep+Dasari%2C+Joey+Hejna%2C+Charles+Xu%2C+Jianlan+Luo%2C+Tobias+Kreiman%2C+You+Liang+Tan%2C+Pannag+Sanketi%2C+Quan+Vuong%2C+Ted+Xiao)
52. Open X-Embodiment Collaboration, Abhishek Padalkar, Acorn Pooley, Ajinkya Jain, Alex Bewley, Alex Herzog, Alex Irpan, Alexander Khazatsky, Anant Rai, Anikait Singh, Anthony Brohan, Antonin Raffin, Ayzaan Wahid, Ben Burgess-Limerick, Beomjoon Kim, Bernhard Schölkopf, Brian Ichter, Cewu Lu, Charles X… [find](https://scholar.google.com/scholar?q=Open+X%2DEmbodiment+Collaboration%2C+Abhishek+Padalkar%2C+Acorn+Pooley%2C+Ajinkya+Jain%2C+Alex+Bewley%2C+Alex+Herzog%2C+Alex+Irpan%2C+Alexander+Khazatsky%2C+Anant+Rai%2C+Anikait+Singh%2C+Anthony+Brohan%2C+Antonin+Raffin%2C+Ayz)
53. Artidoro Pagnoni, Ram Pasunuru, Pedro Rodriguez, John Nguyen, Benjamin Muller, Margaret Li, Chunting Zhou, Lili Yu, Jason Weston, Luke Zettlemoyer, Gargi Ghosh, Mike Lewis, Ari Holtzman†, and Srinivasan Iyer. Byte latent transformer: Patches scale better than tokens. 2024. URL [https://github.com/fac…](https://github.com/fac…) [↗](https://github.com/fac…)
54. Haozhi Qi, Ashish Kumar, Roberto Calandra, Yi Ma, and Jitendra Malik. In-hand object rotation via rapid motor adaptation, 2022. URL https://arxiv.org/abs/2210.04887. — [arXiv:2210.04887](https://arxiv.org/abs/2210.04887) [↗](https://arxiv.org/abs/2210.04887)
55. Alec Radford, Jeff Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. Language models are unsupervised multitask learners. 2019. [find](https://scholar.google.com/scholar?q=Alec+Radford%2C+Jeff+Wu%2C+Rewon+Child%2C+David+Luan%2C+Dario+Amodei%2C+and+Ilya+Sutskever%2E+Language+models+are+unsupervised+multitask+learners%2E+2019%2E)
56. Scott Reed, Konrad Zolna, Emilio Parisotto, Sergio Gómez Colmenarejo, Alexander Novikov, Gabriel Barth-maron, Mai Giménez, Yury Sulsky, Jackie Kay, Jost Tobias Springenberg, et al. A generalist agent. Transactions on Machine Learning Research, 2022. [find](https://scholar.google.com/scholar?q=Scott+Reed%2C+Konrad+Zolna%2C+Emilio+Parisotto%2C+Sergio+Go%CC%81mez+Colmenarejo%2C+Alexander+Novikov%2C+Gabriel+Barth%2Dmaron%2C+Mai+Gime%CC%81nez%2C+Yury+Sulsky%2C+Jackie+Kay%2C+Jost+Tobias+Springenberg%2C+et+al%2E+A+generalist+agen)
57. Rico Sennrich, Barry Haddow, and Alexandra Birch. Neural machine translation of rare words with subword units. arXiv preprint arXiv:1508.07909, 2015. [our card](https://md.research.elamur.ai/free/en/papers/sennrich-2016-neural-machine-translation-rare-words/)
58. Himanshu Gaurav Singh, Antonio Loquercio, Carmelo Sferrazza, Jane Wu, Haozhi Qi, Pieter Abbeel, and Jitendra Malik. Hand-object interaction pretraining from videos, 2024. URL https://arxiv.org/abs/2409.08273. — [arXiv:2409.08273](https://arxiv.org/abs/2409.08273) [↗](https://arxiv.org/abs/2409.08273)
59. Aaron van den Oord, Oriol Vinyals, and Koray Kavukcuoglu. Neural discrete representation learning, 2018. URL https://arxiv.org/abs/1711.00937. — [arXiv:1711.00937](https://arxiv.org/abs/1711.00937) [↗](https://arxiv.org/abs/1711.00937)
60. Homer Rich Walke, Kevin Black, Tony Z Zhao, Quan Vuong, Chongyi Zheng, Philippe Hansen-Estruch, Andre Wang He, Vivek Myers, Moo Jin Kim, Max Du, et al. BridgeData v2: A dataset for robot learning at scale. In Conference on Robot Learning, pages 1723– 1736. PMLR, 2023. [find](https://scholar.google.com/scholar?q=Homer+Rich+Walke%2C+Kevin+Black%2C+Tony+Z+Zhao%2C+Quan+Vuong%2C+Chongyi+Zheng%2C+Philippe+Hansen%2DEstruch%2C+Andre+Wang+He%2C+Vivek+Myers%2C+Moo+Jin+Kim%2C+Max+Du%2C+et+al%2E+BridgeData+v2%3A+A+dataset+for+robot+learning+at+s)
61. Gregory K Wallace. The jpeg still picture compression standard. IEEE transactions on consumer electronics, 38 (1):xviii–xxxiv, 1992. [find](https://scholar.google.com/scholar?q=Gregory+K+Wallace%2E+The+jpeg+still+picture+compression+standard%2E+IEEE+transactions+on+consumer+electronics%2C+38+%281%29%3Axviii%E2%80%93xxxiv%2C+1992%2E)
62. Lirui Wang, Xinlei Chen, Jialiang Zhao, and Kaiming He. Scaling proprioceptive-visual learning with heterogeneous pre-trained transformers. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. [find](https://scholar.google.com/scholar?q=Lirui+Wang%2C+Xinlei+Chen%2C+Jialiang+Zhao%2C+and+Kaiming+He%2E+Scaling+proprioceptive%2Dvisual+learning+with+heterogeneous+pre%2Dtrained+transformers%2E+In+The+Thirty%2Deighth+Annual+Conference+on+Neural+Information)
63. Junjie Wen, Yichen Zhu, Jinming Li, Minjie Zhu, Kun Wu, Zhiyuan Xu, Ning Liu, Ran Cheng, Chaomin Shen, Yaxin Peng, Feifei Feng, and Jian Tang. Tinyvla: Towards fast, data-efficient vision-language-action models for robotic manipulation. arXiv preprint arXiv:2409.12514, 2024. — [arXiv:2409.12514](https://arxiv.org/abs/2409.12514) [find](https://scholar.google.com/scholar?q=Junjie+Wen%2C+Yichen+Zhu%2C+Jinming+Li%2C+Minjie+Zhu%2C+Kun+Wu%2C+Zhiyuan+Xu%2C+Ning+Liu%2C+Ran+Cheng%2C+Chaomin+Shen%2C+Yaxin+Peng%2C+Feifei+Feng%2C+and+Jian+Tang%2E+Tinyvla%3A+Towards+fast%2C+data%2Defficient+vision%2Dlanguage%2Dact)
64. Wilson Yan, Matei Zaharia, Volodymyr Mnih, Pieter Abbeel, Aleksandra Faust, and Hao Liu. Elastictok: Adaptive tokenization for image and video. arXiv preprint arXiv:2410.08368, 2024. — [arXiv:2410.08368](https://arxiv.org/abs/2410.08368) [find](https://scholar.google.com/scholar?q=Wilson+Yan%2C+Matei+Zaharia%2C+Volodymyr+Mnih%2C+Pieter+Abbeel%2C+Aleksandra+Faust%2C+and+Hao+Liu%2E+Elastictok%3A+Adaptive+tokenization+for+image+and+video%2E+arXiv+preprint+arXiv%3A2410%2E08368%2C+2024%2E+%E2%80%94+%5BarXiv%3A2410%2E083)
65. Seonghyeon Ye, Joel Jang, Byeongguk Jeon, Sejune Joo, Jianwei Yang, Baolin Peng, Ajay Mandlekar, Reuben Tan, Yu-Wei Chao, Bill Yuchen Lin, et al. Latent action pretraining from videos. arXiv preprint arXiv:2410.11758, 2024. — [arXiv:2410.11758](https://arxiv.org/abs/2410.11758) [find](https://scholar.google.com/scholar?q=Seonghyeon+Ye%2C+Joel+Jang%2C+Byeongguk+Jeon%2C+Sejune+Joo%2C+Jianwei+Yang%2C+Baolin+Peng%2C+Ajay+Mandlekar%2C+Reuben+Tan%2C+Yu%2DWei+Chao%2C+Bill+Yuchen+Lin%2C+et+al%2E+Latent+action+pretraining+from+videos%2E+arXiv+preprint+)
66. Lijun Yu, Yong Cheng, Kihyuk Sohn, José Lezama, Han Zhang, Huiwen Chang, Alexander G. Hauptmann, Ming-Hsuan Yang, Yuan Hao, Irfan Essa, and Lu Jiang. Magvit: Masked generative video transformer, 2023. URL https://arxiv.org/abs/2212.05199. — [arXiv:2212.05199](https://arxiv.org/abs/2212.05199) [↗](https://arxiv.org/abs/2212.05199)
67. Michał Zawalski, William Chen, Karl Pertsch, Oier Mees, Chelsea Finn, and Sergey Levine. Robotic control via embodied chain-of-thought reasoning. In Conference on Robot Learning, 2024. [find](https://scholar.google.com/scholar?q=Micha%C5%82+Zawalski%2C+William+Chen%2C+Karl+Pertsch%2C+Oier+Mees%2C+Chelsea+Finn%2C+and+Sergey+Levine%2E+Robotic+control+via+embodied+chain%2Dof%2Dthought+reasoning%2E+In+Conference+on+Robot+Learning%2C+2024%2E)
68. Neil Zeghidour, Alejandro Luebs, Ahmed Omran, Jan Skoglund, and Marco Tagliasacchi. Soundstream: An end-to-end neural audio codec, 2021. URL https://arxiv.org/abs/2107.03312. [↗](https://arxiv.org/abs/2107.03312) [our card](https://md.research.elamur.ai/free/en/papers/zeghidour-2021-soundstream-end-end-neural-audio/)
69. Tony Z Zhao, Vikash Kumar, Sergey Levine, and Chelsea Finn. Learning fine-grained bimanual manipulation with low-cost hardware. arXiv preprint arXiv:2304.13705, 2023. [find](https://scholar.google.com/scholar?q=Tony+Z+Zhao%2C+Vikash+Kumar%2C+Sergey+Levine%2C+and+Chelsea+Finn%2E+Learning+fine%2Dgrained+bimanual+manipulation+with+low%2Dcost+hardware%2E+arXiv+preprint+arXiv%3A2304%2E13705%2C+2023%2E)
70. Tony Z Zhao, Jonathan Tompson, Danny Driess, Pete Florence, Kamyar Ghasemipour, Chelsea Finn, and Ayzaan Wahid. Aloha unleashed: A simple recipe for robot dexterity. arXiv preprint arXiv:2410.13126, 2024. — [arXiv:2410.13126](https://arxiv.org/abs/2410.13126) [find](https://scholar.google.com/scholar?q=Tony+Z+Zhao%2C+Jonathan+Tompson%2C+Danny+Driess%2C+Pete+Florence%2C+Kamyar+Ghasemipour%2C+Chelsea+Finn%2C+and+Ayzaan+Wahid%2E+Aloha+unleashed%3A+A+simple+recipe+for+robot+dexterity%2E+arXiv+preprint+arXiv%3A2410%2E13126%2C+2)
71. Haoyu Zhen, Xiaowen Qiu, Peihao Chen, Jincheng Yang, Xin Yan, Yilun Du, Yining Hong, and Chuang Gan. 3d-vla: A 3d vision-language-action generative world model. arXiv preprint arXiv:2403.09631, 2024. — [arXiv:2403.09631](https://arxiv.org/abs/2403.09631) [find](https://scholar.google.com/scholar?q=Haoyu+Zhen%2C+Xiaowen+Qiu%2C+Peihao+Chen%2C+Jincheng+Yang%2C+Xin+Yan%2C+Yilun+Du%2C+Yining+Hong%2C+and+Chuang+Gan%2E+3d%2Dvla%3A+A+3d+vision%2Dlanguage%2Daction+generative+world+model%2E+arXiv+preprint+arXiv%3A2403%2E09631%2C+2024%2E+)
72. Haoyu Zhen, Xiaowen Qiu, Peihao Chen, Jincheng Yang, Xin Yan, Yilun Du, Yining Hong, and Chuang Gan. 3d-vla: 3d vision-language-action generative world model. arXiv preprint arXiv:2403.09631, 2024. — [arXiv:2403.09631](https://arxiv.org/abs/2403.09631) [find](https://scholar.google.com/scholar?q=Haoyu+Zhen%2C+Xiaowen+Qiu%2C+Peihao+Chen%2C+Jincheng+Yang%2C+Xin+Yan%2C+Yilun+Du%2C+Yining+Hong%2C+and+Chuang+Gan%2E+3d%2Dvla%3A+3d+vision%2Dlanguage%2Daction+generative+world+model%2E+arXiv+preprint+arXiv%3A2403%2E09631%2C+2024%2E+%E2%80%94+)
73. Ruijie Zheng, Yongyuan Liang, Shuaiyi Huang, Jianfeng Gao, Hal Daumé III, Andrey Kolobov, Furong Huang, and Jianwei Yang. Tracevla: Visual trace prompting enhances spatial-temporal awareness for generalist robotic policies. arXiv preprint arXiv:2412.10345, 2024. — [arXiv:2412.10345](https://arxiv.org/abs/2412.10345) [find](https://scholar.google.com/scholar?q=Ruijie+Zheng%2C+Yongyuan+Liang%2C+Shuaiyi+Huang%2C+Jianfeng+Gao%2C+Hal+Daume%CC%81+III%2C+Andrey+Kolobov%2C+Furong+Huang%2C+and+Jianwei+Yang%2E+Tracevla%3A+Visual+trace+prompting+enhances+spatial%2Dtemporal+awareness+for+gene)
74. Zhiyuan Zhou, Pranav Atreya, Abraham Lee, Homer Walke, Oier Mees, and Sergey Levine. Autonomous improvement of instruction following skills via foundation models. In Conference on Robot Learning, 2024. [find](https://scholar.google.com/scholar?q=Zhiyuan+Zhou%2C+Pranav+Atreya%2C+Abraham+Lee%2C+Homer+Walke%2C+Oier+Mees%2C+and+Sergey+Levine%2E+Autonomous+improvement+of+instruction+following+skills+via+foundation+models%2E+In+Conference+on+Robot+Learning%2C+2024)
75. Jacob Ziv and Abraham Lempel. Compression of individual sequences via variable-rate coding. IEEE transactions on Information Theory, 24(5):530–536, 1978. A PPENDIX [find](https://scholar.google.com/scholar?q=Jacob+Ziv+and+Abraham+Lempel%2E+Compression+of+individual+sequences+via+variable%2Drate+coding%2E+IEEE+transactions+on+Information+Theory%2C+24%285%29%3A530%E2%80%93536%2C+1978%2E+A+PPENDIX)