Jian Yang

LLM Researcher in Code Large Language Model

[Jian Yang @jian_yang96]

3/ Result:
On 21 sources / 50B tokens:
👉 matches full-corpus performance 👉 using only 25B tokens (50% data)
Paper link：https://t.co/CawcCOKCuO

[Jian Yang @jian_yang96]

1/
We propose MIRA, a framework for mid-training data selection.
Core idea: 👉 learn source-specific quality rubrics instead of using one global scorer 👉 distill them into scalable scoring models
This makes data selection adaptive to data type, not just sample-level ranking.

[Jian Yang @jian_yang96]

2/ Why this matters:
Mid-training data is highly heterogeneous:
code / QA / agents / docs
A single quality function fails to capture:
👉 different formats
👉 different capability signals
👉 different “goodness” definitions

[Jian Yang @jian_yang96]

Welcome to INLG 2026 !

[INLG 2026 @inlgmeeting]

The Call for Papers for #INLG2026 is out!

🗓️ Submit by July 15 (AoE)
💍 ARR commit by August 5
🆕 Squibs welcomed (raising an issue without needing to solve it)
🆕 Non-archival track for WIP
📍Utrecht, NL — Oct 17–21, just before EMNLP

https://t.co/eLnNE3Fqrc
#NLProc #INLG

[Jian Yang @jian_yang96]

Maybe Claude looped many times 🐶, but our research shows a 7B model needs Only Loop Once to rival giants—because the 2nd pass is the ceiling and the 3rd is a cliff dive 📉🚀.

[Yaxin Du @DorothyDDU]

LoopCoder-v2 is out 🚀

Loop Transformers reuse the same block for recurrent hidden-state refinement — letting models “think” more without simply stacking more layers.

We study how many loops are actually worth it in Parallel Loop Transformers.

📜https://t.co/6iKMPelqQ2

[Qian Liu @sivil_taram]

🔍 From simple code completion to autonomous software engineering agents — what changed in the past 5 years?

We wrote the playbook 📖 "𝐅𝐫𝐨𝐦 𝐂𝐨𝐝𝐞 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐌𝐨𝐝𝐞𝐥𝐬 𝐭𝐨 𝐀𝐠𝐞𝐧𝐭𝐬" — 300 pages covering exact recipes 🧪, scaling laws 📈 & RL techniques 🎯 for state-of-the-art Code LLMs.

What's inside:
✨ Full lifecycle: Data → Pre-training → SFT → RL specifically for Code LLMs.

🧪 Empirical Training Recipes: We reveal language-specific scaling laws (Python vs. Java), where Python benefits massively from scale, but C# and Java are "easier" to learn and saturate faster.

🤖 SWE Agents Taxonomy: A detailed look at agents that handle the real tasks, including Environment, Dev, Testing, and Maintenance — moving beyond simple generation to full workflow automation.

This work was led by Jian Yang (@jian_yang96 ) at Beihang University, alongside a stellar collaboration of researchers from Alibaba, ByteDance, and other affiliations. And please find more details in the paper!

🔗 https://t.co/sJwikNRw7H