Minghao Yan

PhDing @WisconsinCS | prev @RiceCompSci, @Google, @AWS, @ThirdAILab (acq. by @ServiceNow), avid fan of @Uruguay

[Minghao Yan @Minghao__Yan]

🚀🚀🚀

[Omenn-Darling Bioengineering Institute @omenndarlingbio]

Scientific American has named ODBI’s Kaiyi Jiang to its inaugural list of Young American Scientists. Congrats, @idmjky!

Jiang will join ODBI next month. Read more on the ODBI site: https://t.co/nziKegQQqm

[Minghao Yan @Minghao__Yan]

@awsTO USMNT plays like how Canada sees itself in the mirror 🤣

[Hongyi Wang @HongyiWang10]

1/8) Excited to share PR²: Predictive Routing Replay for MoE-Based LLM RL! 🎉

This is also the first paper from our research group at @RutgersCS @RutgersU.

While MoE LLMs scale remarkably well, RL training exposes a hidden source of instability: routing.

Paper: https://t.co/uekygKHSON

A special shout-out to my talented PhD student, @DaizeDongCS, for leading this project and driving many of its key ideas and experiments.

[Minghao Yan @Minghao__Yan]

@ChengleiSi Hard agree! If NanoGPT speedrun and Karpathy’s auto research has taught us anything, it’s that we still need major breakthrough(s) for agents to discover paradigm shift ideas.

[Minghao Yan @Minghao__Yan]

Spot on! One of my biggest takeaways after working on AI scientists for a while.

[Andrew Gordon Wilson @andrewgwils]

Being good at next word prediction is the opposite of what we want for creativity, for scientific breakthroughs.

[Minghao Yan @Minghao__Yan]

@karpathy We worked on building an evolutionary agent for the NanoGPT benchmark back in October and shared our findings in the paper: https://t.co/C8CxWBIGKH

Similarly, we also observed that the agent is really good at tuning hyperparameters, designing context / lr / decay schedules!

[Minghao Yan @Minghao__Yan]

@_ScottCondron @karpathy Shameless self-plug here: we’ve worked on the very task of self-evolution on the NanoGPT benchmark in our paper!

https://t.co/y0jMrWNyni

[Minghao Yan @Minghao__Yan]

We even deployed PACEvolve on the Modded NanoGPT challenge. Despite the benchmark being heavily optimized by the community, PACEvolve discovered further gains in data loading, network initialization, and tuned better hyperparameters.

[Minghao Yan @Minghao__Yan]

Code drop here:

[Minghao Yan @Minghao__Yan]

Our code is finally out at https://t.co/2VxiNAXw4f.

Run it with your favorite task and see if you can push the scientific frontier!

[Minghao Yan @Minghao__Yan]

🚀 Thrilled to introduce PACEvolve: Enabling Long-Horizon Progress-Aware Consistent Evolution.
We show how to push LLM self-evolution beyond short, unstable improvements and into consistent, long-horizon gains. 🧵👇

[Minghao Yan @Minghao__Yan]

This work was done during my internship at Google and would not have been possible without my mentors and collaborators across Google and DeepMind. Kudos to everyone involved!
Paper: https://t.co/yYL82IjYFx
Code drop coming soon, stay tuned!

[Minghao Yan @Minghao__Yan]

@karpathy Yes, we’ve been working on building a RSI agent on NanoGPT speedup since last Oct! Check it out in our paper:

https://t.co/y0jMrWNyni

[Minghao Yan @Minghao__Yan]

We even deployed PACEvolve on the Modded NanoGPT challenge. Despite the benchmark being heavily optimized by the community, PACEvolve discovered further gains in data loading, network initialization, and tuned better hyperparameters.

[Minghao Yan @Minghao__Yan]

@eliebakouch agree! we tried it in our paper and it has been an awesome experience learning about both the capabilities and the limitations of current frontier LLMs.

[Minghao Yan @Minghao__Yan]

We even deployed PACEvolve on the Modded NanoGPT challenge. Despite the benchmark being heavily optimized by the community, PACEvolve discovered further gains in data loading, network initialization, and tuned better hyperparameters.

[Minghao Yan @Minghao__Yan]

Our code is finally out at https://t.co/2VxiNAXw4f.

Run it with your favorite task and see if you can push the scientific frontier!

[Minghao Yan @Minghao__Yan]

🚀 Thrilled to introduce PACEvolve: Enabling Long-Horizon Progress-Aware Consistent Evolution.
We show how to push LLM self-evolution beyond short, unstable improvements and into consistent, long-horizon gains. 🧵👇

[Henry Shevlin @dioscuri]

All I want for Christmas is a new Matt Lakeman blogpost