Skip to content

Author

Zi-Ming Liu

We have 3 of 3 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Preprint Oct 2026

Rethinking Activation Functions: Reducing Bias Can Improve Learning

Many widely used activation functions, including ReLU, GELU, Swish, and Sigmoid, produce predominantly positive outputs, resulting in a nonzero post-activation mean that we term activation bias. Despite its ubiquity, the effect of activation bias on learning dynamics remains poorly understood. Using controlled models,...

Zi-Rui Peng, Helen Zheng, Zi-Ming Liu · 0 citations
#machine learning Preprint Sep 2026

Beyond Quadratic Loss: The Stability Phase Diagram of Adam

This work connects Adam loss spikes to both the mismatch between momentum timescales and finite-scale superquadratic loss geometry beyond the Hessian, and shows that confident cross-entropy losses develop a core--wall landscape comprising a narrow quadratic core followed by a steep wall, which produces effective superq...

Gao-Xiang Tang, Huan-Ran Chen, Zi-Ming Liu · 0 citations
#machine learning Preprint Sep 2026

A Chosen Future Can Still Be Rewritten: Causal Writability in Video Models

It is shown that when a video model generates physically incorrect motion, the correct motion remains available inside the model and can still be made to control the generated video, and a stronger downstream write can restore physical motion, while excessive gain overshoots.

Xing-Yun Wang, Hao-Min Zheng, Man Yuan et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.