arXiv AI

LaMoC: Loss-Aware Modular Compression for LLMs

arXiv Computer Vision
Aug 27

SHIFT-LLM: Distribution Shift Correction in Depth-Pruned LLMs

SHIFT-LLM is a training‑free post‑pruning correction framework that inserts a Linear Residual Adapter (LRA) at each depth‑pruned site in large language models. Each LRA preserves the original residual identity while adding a lightweight affine correction calibrated via closed‑form least‑squares regression on a small held‑out set, thereby approximating the hidden state that would have been produced by the removed block. Experiments across multiple model families and benchmarks show that SHIFT‑LLM consistently recovers accuracy lost to depth pruning, achieving gains up to +15.7 points on Llama‑3.1‑8B‑Instruct with only a few hundred calibration samples and no gradient computation.

By Ali Bahri, Hang Li, Hongliang Li, Zhitang Chen
arXiv AI
Sep 3

Scalable Kronecker-Fisher Approximation: Efficient Hessian Analysis for Billion-Parameter Language Models Compression

The paper introduces a scalable Kronecker-based approximation that captures cross-layer interactions without storing the full Fisher matrix, making Hessian analysis feasible for billion-parameter language models. It identifies consistent vulnerability patterns, notably that value projection layers are the most sensitive and exhibit strong cross-layer correlations across various model families. Experiments on quantization, sparsification, inter-layer corruption, and fine-tuning show that the approximation correlates strongly with performance degradation and recovery, providing a practical tool for identifying fragile components and guiding compression and optimization strategies.

By Viacheslav Yusupov, Daria Cherniuk, Evgeny Frolov
arXiv AI
Oct 2

ITC-MoE: Importance-guided Token-aware Compression for MoE Diffusion Language Models

ITC-MoE introduces an Importance-guided Token-aware Compression framework for Mixture-of-Experts Diffusion Language Models. It combines Adaptive Tucker Compression, which uses activation and gradient importance to jointly factorize expert weights and allocate ranks, with Token-aware Compensation and Routing that applies low‑rank adjustments to hot tokens and limits expert candidates for cold tokens. The method achieves significant reductions in computation and storage while maintaining generation quality, exemplified by a 30% compression budget that preserves 96.33% accuracy on MultiArith and delivers up to a 7.22× speedup.

By Lianjun Liu, Shipeng Li, You Huang, Weiqi Yan, Mingte Qiu, Huazhong Liu, Xiaofeng Zhu, Yunshan Zhong
arXiv Machine Learning
Jul 21

NIRVANA: Structured Pruning Reimagined for Large Language Model Compression

arXiv:2509. 14230v2 Announce Type: replace Abstract: While structured pruning presents a highly effective pathway for accelerating Large Language Model (LLM) inference, existing methods frequently suffer from significant performance degradation and demand computationally retraining to recover capabilities.

By Mengting Ai, Tianxin Wei, Sirui Chen, Jingrui He
arXiv Machine Learning
Aug 10

Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models

arXiv:2608. 06901v1 Announce Type: cross Abstract: Vision-language models (VLMs) have achieved remarkable generalization across diverse multimodal tasks through large-scale pre-training, yet their rapidly increasing computational and memory requirements pose significant challenges for deployment in constrained environments.

By Minseok Kang, Hyunwoo Kim, Chanyoung Kim, Minwoo Kim, Jaekoo Lee, Dahuin Jung
arXiv Machine Learning
6d ago

Beyond local reconstruction: Geometry-preserving output alignment for binary LLM quantization

The paper introduces a geometry-preserving output alignment framework for 1‑bit post‑training quantization of large language models. It formulates output alignment as a constrained optimization that balances reconstruction quality with token similarity preservation, and proposes a geometry‑aware salient‑mask selection strategy to choose quantization settings that maintain token relations. Experiments on various LLM families and downstream tasks show consistent gains over existing 1‑bit PTQ methods.

By Dung Anh Hoang, Cuong Pham, Cuong Nguyen, Trung le, Jianfei Cai, Thanh-Toan Do
arXiv Computation and Language
Aug 31

Pruning Laws for Large Language Models

arXiv:2504.04342v2 Announce Type: replace Abstract: Scaling up model parameters and training data consistently improves the performance of large language models (LLMs), but at the cost of rapidly gro...

By Ayan Sengupta, Siddhant Chaudhary, Tanmoy Chakraborty