/ReST-EM-pytorch

Implementations and explorations into the ReST𝐸𝑀 algorithm in the new deepmind paper "Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models"

MIT LicenseMIT

ReST^EM - Pytorch (wip)

Implementations and explorations into the ReST𝐸𝑀 algorithm in the new deepmind paper Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models

Appreciation

Citations

@article{Singh2023BeyondHD,
    title    = {Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models},
    author   = {Avi Singh and John D. Co-Reyes and Rishabh Agarwal and Ankesh Anand and Piyush Patil and Peter J. Liu and James Harrison and Jaehoon Lee and Kelvin Xu and Aaron Parisi and Abhishek Kumar and Alex Alemi and Alex Rizkowsky and Azade Nova and Ben Adlam and Bernd Bohnet and Hanie Sedghi and Igor Mordatch and Isabelle Simpson and Izzeddin Gur and Jasper Snoek and Jeffrey Pennington and Jiri Hron and Kathleen Kenealy and Kevin Swersky and Kshiteej Mahajan and Laura Culp and Lechao Xiao and Maxwell L. Bileschi and Noah Constant and Roman Novak and Rosanne Liu and Tris Brian Warkentin and Yundi Qian and Ethan Dyer and Behnam Neyshabur and Jascha Narain Sohl-Dickstein and Noah Fiedel},
    journal  = {ArXiv},
    year     = {2023},
    volume   = {abs/2312.06585},
    url      = {https://api.semanticscholar.org/CorpusID:266163375}
}