Author

Sepehr Seifi

1 paper indexed here

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.

#machine learning Feb 2026

Zero Sum SVD: Balancing Loss Sensitivity for Low Rank LLM Compression

Zero Sum SVD (ZS-SVD) is proposed, a post-training method that performs global singular component selection using activation whitening and first-order calibration loss estimates in whitened coordinates, automatically yielding heterogeneous ranks without solving a rank allocation optimization.

Ali Abbasi, Chayne Thrash, Hao-Ran Qin et al. · arXiv.org · 9 citations · ⚡1