IrekoGPT: Turning Structured Pruning into Post-Hoc Slimmable LLMs

Abstract:We introduce IrekoGPT, a post-hoc method for converting pretrained LLMs into slimmable models whose width can be adjusted at inference time. Building on SliceGPT, we retain its projection matrices without pruning them, allowing a single model to expose nested subnetworks at different widths. We improve robustness by calibrating each layer across multiple compression ratios, and correct downstream linear layers through gradient-free ridge regression. Across Llama and Qwen models, preliminary results show improvements over naive PCA-based slimming, with the largest gains at high compression. Code is available at this https URL

Accepted at the NeurIPS 2026 Workshop "AXIOM: Foundations of Efficient Deep Learning". 8 pages, 5 figures

Machine Learning (cs.LG); Machine Learning (stat.ML)

arXiv:2610.00426 [cs.LG]

(or arXiv:2610.00426v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2610.00426 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.