Model Releases
IrekoGPT: Turning Structured Pruning into Post-Hoc Slimmable LLMs
arXiv:2610.00426v1 Announce Type: new 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
arXiv:2610.00426v1 Announce Type: new 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 https://github.com/aimagelab/IrekoGPT
Related
- COEC: Calibrated Orthogonal-Equivalence Compensation for Structured Pruning of Large Language Models
- Damage-Aware Bandit Pruning for Vision and Language Transformers
- DarwinLM: Evolutionary Structured Pruning of Large Language Models
- GroupMask: Layer-Adaptive Group-wise Sparsity for Semi-Structured LLM Pruning
Source: arXiv cs.LG | 2026-10-02