On Mesa-Optimization in Autoregressively Trained Transformers: Emergence and Capability

Item #:

079017-1555

Details

DOI: https://doi.org/10.52202/079017-1555
Author(s): Wei Huang, Chongxuan Li, Rongzhen Wang, Guoqiang Wu, Chenyu Zheng, Jun Zhu
Pages: 49081-49129 (49 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 37
Date/Location: Held 10-15 December 2024, Vancouver, Canada.
Series: Advances in Neural Information Processing Systems 37
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)

DOI: https://doi.org/10.52202/079017-1555
Author(s): Wei Huang, Chongxuan Li, Rongzhen Wang, Guoqiang Wu, Chenyu Zheng, Jun Zhu
Pages: 49081-49129 (49 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 37
Date/Location: Held 10-15 December 2024, Vancouver, Canada.
Series: Advances in Neural Information Processing Systems 37
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)