HET, a sparse large model training acceleration program cooperating with Tencent and Peking University, was selected into the international top conference VLDB

Others 2022-04-20 06:21:08 views: 0

NoSuchKey

Guess you like

Origin http://43.154.161.224:23101/article/api/json?id=324140688&siteId=291194637

HET, a sparse large model training acceleration program cooperating with Tencent and Peking University, was selected into the international top conference VLDB

good news! Two papers from the ByteDance infrastructure computing team were selected for VLDB, the top database conference

Interpretation of the paper at the top database conference VLDB 2023: How ByteDance solves the problem of operation and maintenance of ultra-large-scale streaming tasks

CloudWeGo - Sonic was selected into the graduate program of Peking University, opening a new chapter of school-enterprise cooperation

Tencent Youtu was selected as the top artificial intelligence conference AAAI papers -- 10 papers

Peking University officially released ChatLaw, a large Chinese legal model, and made it open source

CodeShell, a large open source code model of Peking University, provides supporting IDE plug-ins

Flink OLAP helps ByteHTAP debut in VLDB, the top database conference

LDA* based on Tencent Angel was selected into VLDB, surpassing Microsoft LightLDA

Alibaba open source large-scale sparse model training/prediction engine DeepRec

PLDI'23 | June 17-21, the top conference in the field of programming languages, scholars from Peking University, Nanjing University, and Zhejiang University serve as the Review Committee!

The top international academic conference ISSTA was held, and Sun Yat-sen University and WeBank jointly published the latest research results of blockchain

The server was overwhelmed, and ChatLaw, a large legal model of Peking University, became popular: directly tell you how Zhang San was sentenced!

ICML 2023 Outstanding Papers Reduced to 6 Substantially! Alumni of Peking University and Wuhan Institute of Technology won awards, and large model watermarks were favored

Peking University used ChatGPT to set up a development team: the large model played multiple roles and collaborated to complete software development tasks

2023 World Artificial Intelligence Conference, Hejing Technology was selected into the "2023 Large Model and AIGC Industry Map" by China Academy of Information and Communications Technology

CCF Stars Project Wenxin University Tour - NLP and large model teacher training is recruiting heavily

Zhejiang University and Alibaba AZFT next-generation database technology laboratory results of cooperation selected VLDB 2019

S&P'23 | May 22-25, the top conference in the field of information security, 37 domestic papers were selected! Ten articles from Zhejiang University!

Large model training time estimation

DeepSpeed accelerates large model training

Prompt Learning in Large Model Training

Large model generation is accelerated by 2 times! A single GPU can be fine-tuned in a few hours, Peking University School of Mathematics alumni jointly work on open source

MIT recognized Chinese university rankings: Tsinghua University, Shanghai Jiaotong University, Zhejiang University, Peking University top 4, Fudan 7

Large model reinforcement learning reward model training

Peking University’s big legal model ChatLaw is popular! !

Fudan University released the low-memory optimization technology LOMO | It reduces the memory usage of large model training to 10.8%, which is far ahead of DeepSpeed!

Pruning basics and actual combat (3): model pruning and sparse training process

Model Compression and Acceleration Methods for Large-Scale Neural Networks

KubeAI large model inference acceleration practice | Dewu Technology

Recommended

Ranking

go common records

SVN power failure recovery

深入理解Redis集群主从复制原理

【二叉树】左叶子之和

[1] The first basic syntax Detailed Kotlin

Linux Ansible creates tasks and executes them

vmware ubuntu virtual machine boots online courses

Use Nodejs to crawl certain data from the web page and write the crawled data into excel (see the next article for the front-end part and the server-side part)

Principle underlying thread pool

The number of bytes occupied when char[ ] is initialized

Daily

2025-03-22(0)

2025-03-21(0)

2025-03-20(0)

2025-03-19(0)

2025-03-18(0)

2025-03-17(0)

2025-03-16(0)

2025-03-15(0)

2025-03-14(0)

2025-03-13(0)