Dobi-SVD: Differentiable SVD for LLM Compression and Some New Perspectives

Qinsi Wang, Jinghan Ke, Masayoshi Tomizuka, Kurt Keutzer, Chenfeng Xu
2/12/2026

No official code found. Best implementation: wangqinsi1/Dobi-SVD (54 stars, MIT, last commit 2025-10-19).

Semantic Scholar

Code Implementations(1)

wangqinsi1/Dobi-SVDCommunity68%
MIT

[ICLR 2025] Dobi-SVD : Differentiable SVD for LLM Compression and Some New Perspectives"

549PythonFeb 6, 20251 years agoMIT

Similar Papers

Auslan-Daily: Australian Sign Language Translation for Daily Communication and News

99% match2026

MMRole: A Comprehensive Framework for Developing and Evaluating Multimodal Role-Playing Agents

99% match1 repo1012026

Spectral-Aware Reservoir Computing for Fast and Accurate Time Series Classification

98% match2026

MG-ViT: A Multi-Granularity Method for Compact and Efficient Vision Transformers

98% match2026

DeeperForward: Enhanced Forward-Forward Training for Deeper and Better Performance

98% match2026

Noise-Assisted Prompt Learning for Image Forgery Detection and Localization

98% match2026

Cite this paper

@article{wang2026dobisvd,
  title  = {Dobi-SVD: Differentiable SVD for LLM Compression and Some New Perspectives},
  author = {Qinsi Wang and Jinghan Ke and Masayoshi Tomizuka and Kurt Keutzer and Chenfeng Xu},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:1959083b64a50b8582aa00452011dbed3e95331a},
  journal = {ICLR 2025 2025}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1959083b64a50b8582aa00452011dbed3e95331a.svg)](https://codeofpaper.com/paper/1959083b64a50b8582aa00452011dbed3e95331a)

Discussion