Výsledky vyhledávání

Akademický článek

Analysis on Airworthiness Special Conditions of Civil Helicopter Fly-by-wire Flight Control System

Autor: MA Liqun, YANG Shibin, SHI Linxuan

Publikováno v: Hangkong gongcheng jinzhan, Vol 12, Iss 3, Pp 20-28 (2021)

The advanced fly-by-wire flight control system can greatly improve the maneuverability and stability of civil helicopters, but this novel design also brings huge challenges to safety assessment.Traditional airworthiness standards may not fully cover

Externí odkaz: https://doaj.org/article/0638860820bc41058b214c8e143953b3

Zobrazit plný text záznamu

Report

FBI-LLM: Scaling Up Fully Binarized LLMs from Scratch via Autoregressive Distillation

Autor: Ma, Liqun, Sun, Mingjie, Shen, Zhiqiang

This work presents a Fully BInarized Large Language Model (FBI-LLM), demonstrating for the first time how to train a large-scale binary language model from scratch (not the partial binary or ternary LLM like BitNet b1.58) to match the performance of

Externí odkaz: http://arxiv.org/abs/2407.07093

Zobrazit plný text záznamu

Akademický článek

The oxidative coupling between benzaldehyde derivatives and phenylacetylene catalyzed by rhodium complexes via C-H bond activation

Autor: Zhao Xinyi, Jia Hongge, Wang Qingji, Song Heming, Tang Yanan, Ma Liqun, Shi Yongqiang, Yang Guoxing, Wang Yazhen, Zang Yu, Xu Shuangping

Publikováno v: Heterocyclic Communications, Vol 26, Iss 1, Pp 20-25 (2020)

This paper reports the use of rhodium (Rh) catalysts for the oxidative coupling reaction between phenylacetylene and benzaldehyde derivatives via C-H bond activation. These reactions were catalyzed by Rh(l-amino acid)(cod) (the l-amino acid is l-phen

Externí odkaz: https://doaj.org/article/cb5337c3138142c9bae676f668c77998

Zobrazit plný text záznamu

Report

Beyond Size: How Gradients Shape Pruning Decisions in Large Language Models

Autor: Das, Rocktim Jyoti, Sun, Mingjie, Ma, Liqun, Shen, Zhiqiang

Large Language Models (LLMs) with billions of parameters are prime targets for network pruning, removing some model weights without hurting performance. Prior approaches such as magnitude pruning, SparseGPT, and Wanda, either concentrated solely on w

Externí odkaz: http://arxiv.org/abs/2311.04902

Zobrazit plný text záznamu

Report

SlimPajama-DC: Understanding Data Combinations for LLM Training

Autor: Shen, Zhiqiang, Tao, Tianhua, Ma, Liqun, Neiswanger, Willie, Liu, Zhengzhong, Wang, Hongyi, Tan, Bowen, Hestness, Joel, Vassilieva, Natalia, Soboleva, Daria, Xing, Eric

This paper aims to understand the impacts of various data combinations (e.g., web text, Wikipedia, GitHub, books) on the pretraining of large language models using SlimPajama. SlimPajama is a rigorously deduplicated, multi-source dataset, which has b

Externí odkaz: http://arxiv.org/abs/2309.10818

Zobrazit plný text záznamu

Report

CSDS: A Fine-Grained Chinese Dataset for Customer Service Dialogue Summarization

Autor: Lin, Haitao, Ma, Liqun, Zhu, Junnan, Xiang, Lu, Zhou, Yu, Zhang, Jiajun, Zong, Chengqing

Dialogue summarization has drawn much attention recently. Especially in the customer service domain, agents could use dialogue summaries to help boost their works by quickly knowing customer's issues and service progress. These applications require s

Externí odkaz: http://arxiv.org/abs/2108.13139

Zobrazit plný text záznamu