We lifted 1 functions out of this paper's own repositories and ran 1 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| Chongjie-Si/AdaMuon | canonical | 1 of 1 |
| Function | Status | Where it lives |
|---|---|---|
| zeropower_via_newtonschulz5 | Ran | Chongjie-Si/AdaMuon/adamuon.py code served (permissive licence) · get_code("5618cdee336eee9f") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
We propose AdaMuon, a novel optimizer that combines element-wise adaptivity with orthogonal updates for large-scale neural network training. AdaMuon incorporates two tightly coupled mechanisms: (1) an element-wise second momentum estimator applied to orthogonalized update directions, and (2) a sign-stabilized orthogonal update, where the momentum is first sign-transformed before orthogonalization. These two components jointly enable variance-adaptive scaling while maintaining stable update geometry. In addition, AdaMuon employs an RMS-aligned rescaling strategy to match the root-mean-square update magnitude to Adam, allowing direct reuse of existing learning rate schedules without extra tuning. Experiments demonstrate that AdaMuon not only maintains stability but can surpass Adam by more than 40\% training efficiency in large-scale scenarios.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2507.11005")
get_code_for_paper("2507.11005")
have("2507.11005")
Connect an agent — have() is free.