Xiaosen Zheng

xszheng2020

2 78 252

https://xszheng2020.github.io

AI & ML interests

Code AI and Data-Centric AI.

Recent Activity

upvoted an article about 6 hours ago

The Most Intuitive Way to Understand Score Function Centering: Understand Trainer-Generator Mismatch in Async RL from the Score Function

liked a model 3 days ago

karpathy/nanochat-d32

upvoted a paper 15 days ago

Code2Video: A Code-centric Paradigm for Educational Video Generation

View all activity

Organizations

upvoted an article about 6 hours ago

Article

The Most Intuitive Way to Understand Score Function Centering: Understand Trainer-Generator Mismatch in Async RL from the Score Function

NormalUhr

•

2 days ago

• 2

liked a model 3 days ago

karpathy/nanochat-d32

Updated Oct 16, 2025 • 6.57k • 96

upvoted a paper 15 days ago

Code2Video: A Code-centric Paradigm for Educational Video Generation

Paper • 2510.01174 • Published Oct 1, 2025 • 36

liked a dataset about 2 months ago

marin-community/grug-moe-mix-swarm

Viewer • Updated Sep 12 • 2.87k • 310 • 1

liked 2 datasets 3 months ago

open-alchemy/code-alchemy

Viewer • Updated Jul 19 • 162M • 2.83k • 17

ByteDance/FullStackBench

Viewer • Updated Dec 4, 2024 • 3.37k • 434 • 30

liked a model 3 months ago

tencent/Hy3

Text Generation • 299B • Updated Jul 20 • 718k • • 982

upvoted a paper 4 months ago

CausalMix: Data Mixture as Causal Inference for Language Model Training

Paper • 2607.01104 • Published Jul 1 • 20

liked a dataset 4 months ago

nvidia/Nemotron-Pretraining-Code-v2

Viewer • Updated Dec 22, 2025 • 836M • 3.9k • 136

liked a model 5 months ago

nvidia/nemotron-climb-fasttext-classifiers

Updated May 11 • 14

liked a dataset 5 months ago

princeton-nlp/SWE-bench_oracle

Viewer • Updated Apr 15, 2024 • 21.5k • 903 • 21

liked a model 5 months ago

ricdomolm/mini-coder-1.7b

Text Generation • 2B • Updated Jun 23 • 370 • 8

liked 2 datasets 5 months ago

AlienKevin/SWE-ZERO-12M-trajectories

Viewer • Updated May 14 • 12.3M • 3.25k • 130

SWE-bench/SWE-bench

Viewer • Updated Aug 16 • 2.52k • 20.9k • 40

upvoted a paper 7 months ago

MiroThinker-1.7 & H1: Towards Heavy-Duty Research Agents via Verification

Paper • 2603.15726 • Published Mar 16 • 188

liked 2 models 7 months ago

miromind-ai/MiroThinker-1.7-mini

Text Generation • 31B • Updated Apr 13 • 1.17k • 104

miromind-ai/MiroThinker-1.7

Text Generation • 235B • Updated Apr 13 • 78 • 145

liked 3 datasets 9 months ago

Xiaosen Zheng

AI & ML interests

Recent Activity

Organizations

xszheng2020's activity

The Most Intuitive Way to Understand Score Function Centering: Understand Trainer-Generator Mismatch in Async RL from the Score Function