Zhang, Shuhao and Yu, Jiayi and Li, Jiarui (2025) Stable Random Sampling (SRS): A New Method to Refine Causal Masking in Decoder-Only Transformer. In: Proceedings of the 2nd International Conference on Machine Learning and Automation, CONF-MLA 2024, November 21, 2024, Adana, Turkey.
![[thumbnail of 71848.pdf]](http://eprints.eai.eu/style/images/fileicons/application_pdf.png "71848.pdf")
PDF
71848.pdf
Download (967kB)
Abstract
In current language modelling, the decoder-only Transformer architecture with causal masking has become a cornerstone, demonstrating exceptional performance across various tasks. However, we have identified two significant limitations: First, causal masking presents a substantial obstacle to further
| Item Type: | Conference or Workshop Item (UNSPECIFIED) |
|---|---|
| Date Deposited: | 04 Mar 2026 18:27 |
| Last Modified: | 16 Apr 2026 21:39 |
| URI: | http://eprints.eai.eu/id/eprint/52528 |
Actions (login required)
- View Item