#234 скачать в хорошем качестве

#234

Foundation models are applied in a broad spectrum of settings with different inference constraints, from massive multi-accelerator clusters to resource-constrained standalone mobile devices. However, the substantial costs associated with training these models often limit the number of unique model sizes that can be offered. Consequently, practitioners are compelled to select a model that may not be optimally aligned with their specific latency and cost requirements. MatFormer is a novel Transformer architecture designed to provide elastic inference across diverse deployment constraints. MatFormer achieves this by incorporating a nested Feed Forward Network (FFN) block structure within a standard Transformer model. During training, the parameters of multiple nested FFN blocks are optimized with varying sizes, enabling the extraction of hundreds of accurate smaller models without incurring additional computational costs. Efficacy of MatFormer is validated across different model classes (decoders and encoders) and modalities (language and vision), demonstrating its potential for real-world deployment. A 850M decoder-only MatFormer language model (MatLM) allows us to extract multiple smaller models spanning from 582M to 850M parameters, each exhibiting better validation loss and one-shot downstream evaluations than independently trained counterparts. Furthermore, smaller encoders extracted from a universal MatFormer-based ViT (MatViT) encoder preserve the metric-space structure for adaptive large-scale retrieval. Finally, speculative decoding with the accurate and consistent submodels extracted from MatFormer can lead to significant reduction in inference latency. In this video, I talk about the following: How are the MatFormer models trained? How does MatFormer perform? For more details, please look at https://arxiv.org/pdf/2310.07707 and https://github.com/devvrit/matformer Kudugunta, Sneha, Aditya Kusupati, Tim Dettmers, Kaifeng Chen, Inderjit Dhillon, Yulia Tsvetkov, Hannaneh Hajishirzi, Sham Kakade, Ali Farhadi, and Prateek Jain. "Matformer: Nested transformer for elastic inference." arXiv preprint arXiv:2310.07707 (2023).