Yang Liu

@nlpyang

#LLM Researcher @Microsoft; PhD @EdinburghNLP

Bellevue, WA

nlp-yang.github.io

Replies Reposts Media

Pinned
Yang Liu
@nlpyang
Jun 3
Excited to introduce our MAI Code model at Microsoft Build. As shared in the session, this is a MoE (5B active / 137B total) initialized from an MAI pretrained model and trained for real user scenarios with product harnesses. I’m proud to have served as the research lead for this
Microsoft AI
@MicrosoftAI
Jun 3
MAI-Code-1-Flash is here! Built and optimized for GitHub Copilot. From quick fixes to complex engineering challenges, write better code with more return on token. Rolling out to GitHub Copilot individual users in Visual Studio Code in the model picker and under the default auto
2
Yang Liu
@nlpyang
Sep 2
Raptor Mini is retiring, it was such a memorable year from zero to this.
Daniel Estrada
@BuiltByEstrada
Sep 1
GitHub Copilot drops six models today across chat, inline edits, agent mode, and completions. Gemini 3.1 Pro, Claude Opus 4.5 and 4.6, Claude Sonnet 4.5, Claude Sonnet 4.6 for most plans, and Raptor Mini are gone. Enterprise admins have to enable the successors in model policy
1
Yang Liu
@nlpyang
Mar 3, 2025
Missing coding data in your R1? 🔥 Introducing KodCode—the largest verified synthetic coding dataset for Code LLM training! • 447K question–solution–test triplets • 12 diverse subsets • 10-trial solution verification for rock-solid correctness kodcode-ai.github.io
3
Yang Liu
@nlpyang
Jan 27, 2025
I don't really understand why people think RL takes less compute than pretrain.
2
Yang Liu
@nlpyang
Aug 21, 2024
Check our paper so you can really challenge an LLM
Yulong Chen
@Yulongchen1010
Aug 20, 2024
Evaluating LLMs usually requires sophisticated human designs and with the continuous improvement of LLMs, it is difficult for humans to find their limitations. Can LLMs find their own limitations by proposing questions to themselves? Check our new paper: arxiv.org/abs/2408.08978