Yukuo Profile Banner
Yukuo Profile
Yukuo

@cenyk1230

Followers
3,038
Following
591
Media
10
Statuses
42

Obtained Ph.D. degree from the DCST in @Tsinghua_Uni , focusing on #ChatGLM () and coding assistant #CodeGeeX .

Beijing
Joined October 2017
Don't wanna be here? Send us removal request.
Explore trending content on Musk Viewer
@cenyk1230
Yukuo
9 months
#ChatGLM3 is a new generation of pre-trained conversational models jointly released by Zhipu AI and KEG Lab of Tsinghua University. 🔗 Github: 🔗 Model (ChatGLM3-6B): ChatGLM3-6B is the open-source model in the ChatGLM3 series.
17
4K
6K
@cenyk1230
Yukuo
8 months
📢 Introducing BPO: Black-Box Prompt Optimization: Aligning #LLMs without Model Training (by #ChatGLM Team) 🚀 📝 Paper: 💻 Code: Not everyone is familiar with how to communicate effectively with large language models (LLMs). One
Tweet media one
Tweet media two
Tweet media three
Tweet media four
70
4K
6K
@cenyk1230
Yukuo
9 months
#ChatGLM3 -Turbo, the third-generation model developed by #Tsinghua & #ZhipuAI , has remarkably set a new record in the SuperCLUE-Agent Chinese benchmark. 🔗 Github: 🔗 Benchmark: This achievement showcases a significant improvement
Tweet media one
Tweet media two
6
4K
6K
@cenyk1230
Yukuo
8 months
📢 Introducing AlignBench, a benchmark specifically designed for Chinese Large Language Models ( #LLMs ), proposed by #ChatGLM Team! It's all about alignment - how well an LLM aligns with human intent, be it following instructions, understanding intent, or producing helpful
Tweet media one
Tweet media two
Tweet media three
Tweet media four
12
4K
6K
@cenyk1230
Yukuo
26 days
🎉Introducing open-source #CodeGeeX4 -ALL-9B, the most powerful, versatile coding model under 10B parameters! The CodeGeeX4 supports code completion, annotation, translation, and advanced features. #Coding #LLMs 🔗Github: 💡Developed by #ZhipuAI , CodeGeeX
Tweet media one
32
149
66
@cenyk1230
Yukuo
9 months
Discover #AgentTuning - a pioneering approach to instruction-tuning of Large Language Models (LLMs) across a variety of tasks. It significantly enhances agent capabilities, exhibiting robust generalization on unseen tasks while preserving core language abilities. The meticulously
Tweet media one
353
150
240
@cenyk1230
Yukuo
9 months
Introducing #CodeGeeX2 , the second-generation multilingual code generation model. It's built on the #ChatGLM architecture, trained on more code data, resulting in a whopping 107% improvement in coding capability over CodeGeeX. 🔗 VSCode plugin: 🔗 Github:
Tweet media one
3
22
62
@cenyk1230
Yukuo
4 years
Our proposed GATNE in KDD 2019 have been implemented by many popular graph toolkits: - Deep Graph Library (DGL): - Paddle Graph Learning (PGL): - CogDL: GATNE paper link: #kdd #kdd2019
0
0
4
@cenyk1230
Yukuo
11 months
Discover #ChatGLM App! Experience the power of Chinese AI at your fingertips. Chat with the app like you would with a friend. Ask questions, get answers, and explore the world of language interaction! #AI #LLM
1
0
3
@cenyk1230
Yukuo
4 years
Our paper "Controllable Multi-Interest Framework for Recommendation" has been accepted to #KDD2020 !! arXiv link: code link:
Tweet media one
Tweet media two
Tweet media three
0
1
2
@cenyk1230
Yukuo
9 months
@labloke11 The ChatGLM3-6B model is publicly available at: . The ChatGLM3-Turbo model can be accessed via API (see: )
1
0
2
@cenyk1230
Yukuo
5 years
Cool! Very interesting attempt to apply BERT to the industrial situation. #BERT #Google
@JeffDean
Jeff Dean (@🏡)
5 years
Great to see this collaboration between Google researchers & engineers launch, with major improvement to search quality! The work brings together many things we've been working on over the last few years: Transformers, BERT, @TensorFlow , TPU pods, ...
8
198
834
0
0
1
@cenyk1230
Yukuo
5 years
@MIT_CSAIL @Karmacoma @jphcoi @mazieres How to combine symbolic methods and neural nets? Maybe symbolic neural nets are the next generation of AI!!!
0
0
1
@cenyk1230
Yukuo
8 months
@yugu_nlp Maybe you can discuss this with Xiao
1
0
1
@cenyk1230
Yukuo
1 year
Physics in a nutshell: it's all about Ψ and how it operates! #QuantumMechanics #Physics #PhysicsFun
@elonmusk
Elon Musk
1 year
Physics is a Ψ op
6K
7K
89K
1
0
1