analiza aiqbench
This research introduces a differentiable Neural Architecture Search framework that jointly optimizes model architecture and mixed-precision quantization, addressing the common limitation of decoupling these processes. By exploring the entire search space rather than constrained subsets, the method achieves up to 1.4x faster inference or 6% higher accuracy on reasoning tasks compared to sequential baselines. This approach offers a significant efficiency gain for deploying pre-trained LLMs on resource-constrained edge devices without requiring extensive retraining.
LLM Compression with Jointly Optimizing Architectural and Quantization choices
arXiv:2606.04063v1 Announce Type: new Abstract: Deploying large language models (LLMs) is challenging due to their significant memory and computational requirements. While some methods address this by developing small or tiny language models from scratch, these approaches demand extensive GPU training. Compressing pre-trained LLMs for edge devices offers a compelling alternative. Beyond pruning and quantization, Neural Architecture Search (NAS) enables effective compression, yet prior NAS approaches often limit the search space and decouple architecture from quantization. We introduce a differentiable NAS framework that explores the entire space and jointly optimizes architectural configurations alongside mixed-precision quantization for linear layers of LLMs. Experiments demonstrate superior accuracy-latency trade-offs: our models achieve up to 1.4x faster inference than sequential NAS-then-quantization baselines at comparable accuracy, or up to 6% higher average accuracy across seven reasoning tasks at equivalent latency.
Abstract:Deploying large language models (LLMs) is challenging due to their significant memory and computational requirements. While some methods address this by developing small or tiny language models from scratch, these approaches demand extensive GPU training. Compressing pre-trained LLMs for edge devices offers a compelling alternative. Beyond pruning and quantization, Neural Architecture Search (NAS) enables effective compression, yet prior NAS approaches often limit the search space and decouple architecture from quantization. We introduce a differentiable NAS framework that explores the entire space and jointly optimizes architectural configurations alongside mixed-precision quantization for linear layers of LLMs. Experiments demonstrate superior accuracy-latency trade-offs: our models achieve up to 1.4x faster inference than sequential NAS-then-quantization baselines at comparable accuracy, or up to 6% higher average accuracy across seven reasoning tasks at equivalent latency.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2606.04063 [cs.LG]
(or arXiv:2606.04063v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.04063 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.
Text integral sub licența deschisă a publicației · 225 cuvinte
Subiecte conexe
Productionizing TurboQuant on AMD GPUs for KV-Cache-Bound LLM Inference
AMD · 11 iun. 2026
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
Latent Space · 17 sept. 2026
[AINews] DeepSeek v4.1-Flash: 763B-P8B-D16B novel causal Encoder–Decoder architecture with vision marks the Return of the Whale
Latent Space · 12 sept. 2026
[AINews] Claude Opus 5.5, the new default model for AINews — and everybody cuts prices 40-50%
Latent Space · 23 sept. 2026