Log inSign up
Ning Ding profile banner
user avatar

Ning Ding

@stingning
Researcher of AI. Assistant Professor @Tsinghua_Uni. Working on scalable methods of language and physical models.
Earth
stingning.cn
Joined May 2015
416
Following
7,192
Followers
RepliesMedia
  • user avatar
    Ning Ding
    @stingning
    Jun 24
    World model has a very clear definition: predicting the next state given the current state and an action. For years that frontier has lived in pixels. This work pushes it into the cyber world, where agents actually operate. Qwen-AgentWorld, led by Yuxin Zuo @zuo_yuxin and Li
    user avatar
    Qwen
    @Alibaba_Qwen
    Jun 24
    📣📣 Meet Qwen-AgentWorld — a native language world model that simulates 7 agent environments (MCP, Search, Terminal, SWE, Web, OS, Android) within a single model. Environment modeling is the training objective from day one, not a post-hoc adaptation. 🤔 LLMs are trained to be
    Image
  • user avatar
    Ning Ding
    @stingning
    Jun 14
    “Months behind” is losing meaning in AI. Time itself has been rescaled — a single month now compresses what used to take years.
  • user avatar
    Ning Ding
    @stingning
    Jun 4
    Happen to have heard of a pretty cool OPD paper: arxiv.org/abs/2604.13016. Recommending for a friend.
    user avatar
    Dwarkesh Patel
    @dwarkesh_sp
    Jun 4
    Recently met @srush_nlp and he started giving me an impromptu lecture on how targeted on-policy self-distillation works. I asked him if I could record it on my iPhone. The basic idea is this: if the model made a mistake at some point in the rollout (for example, calling a tool
    Image
    00:00