Výsledky vyhledávání

Report

InstructionCP: A fast approach to transfer Large Language Models into target language

The rapid development of large language models (LLMs) in recent years has largely focused on English, resulting in models that respond exclusively in English. To adapt these models to other languages, continual pre-training (CP) is often employed, fo

Externí odkaz: http://arxiv.org/abs/2405.20175

Zobrazit plný text záznamu

Report

Chat Vector: A Simple Approach to Equip LLMs with Instruction Following and Model Alignment in New Languages

Autor: Huang, Shih-Cheng, Li, Pin-Zu, Hsu, Yu-Chi, Chen, Kuang-Ming, Lin, Yu Tung, Hsiao, Shih-Kai, Tsai, Richard Tzong-Han, Lee, Hung-yi

Recently, the development of open-source large language models (LLMs) has advanced rapidly. Nevertheless, due to data constraints, the capabilities of most open-source LLMs are primarily focused on English. To address this issue, we introduce the con

Externí odkaz: http://arxiv.org/abs/2310.04799

Zobrazit plný text záznamu

Report

Topological Kondo Superconductors

Autor: Chang, Yung-Yeh, Van Nguyen, Khoe, Chen, Kuang-Lung, Lu, Yen-Wen, Mou, Chung-Yu, Chung, Chung-Hou

Spin-triplet $p$-wave superconductors are promising candidates for topological superconductors. They have been proposed in various heterostructures where a material with strong spin-orbit interaction is coupled to a conventional $s$-wave superconduct

Externí odkaz: http://arxiv.org/abs/2301.00538

Zobrazit plný text záznamu

Report

Compressing Transformer-based self-supervised models for speech processing

Autor: Lin, Tzu-Quan, Yang, Tsung-Huan, Chang, Chun-Yao, Chen, Kuang-Ming, Feng, Tzu-hsun, Lee, Hung-yi, Tang, Hao

Despite the success of Transformers in self- supervised learning with applications to various downstream tasks, the computational cost of training and inference remains a major challenge for applying these models to a wide spectrum of devices. Severa

Externí odkaz: http://arxiv.org/abs/2211.09949

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání