Song Han

Haocheng Xi, Han Cai, Ligeng Zhu et al.

FP8 training has emerged as a promising method for improving training efficiency. Existing frameworks accelerate training by applying FP8 computation to linear layers while leaving optimizer states and activations in higher precision, which fails to fully optimize memory usage. This paper introduces COAT (Compressing Optimizer States and Activations for FP8 Training), a novel FP8 training framework designed to significantly reduce memory footprint when training large models. COAT addresses current limitations through two key innovations: (1) Dynamic Range Expansion, which aligns optimizer state distributions more closely with the FP8 representation range, thereby reducing quantization error, and (2) Mixed-Granularity Activation Quantization, which optimizes activation memory using a combination of per-tensor and per-group quantization strategies. Experiments demonstrate that COAT effectively reduces end-to-end training memory footprint by 1.54x compared to BF16 while achieving nearly lossless performance across various tasks, such as Large Language Model pretraining and fine-tuning and Vision Language Model training. COAT also achieves a 1.43x end-to-end training speedup compared to BF16, performing on par with or surpassing TransformerEngine's speedup. COAT enables efficient full-parameter training of large models on fewer GPUs, and facilitates doubling the batch size in distributed training settings, providing a practical solution for scaling large-scale model training. The code is available at https://github.com/NVlabs/COAT.

Guangxuan Xiao, Tianwei Yin, William T. Freeman et al.

Diffusion models excel at text-to-image generation, especially in subject-driven generation for personalized images. However, existing methods are inefficient due to the subject-specific fine-tuning, which is computationally intensive and hampers efficient deployment. Moreover, existing methods struggle with multi-subject generation as they often blend features among subjects. We present FastComposer which enables efficient, personalized, multi-subject text-to-image generation without fine-tuning. FastComposer uses subject embeddings extracted by an image encoder to augment the generic text conditioning in diffusion models, enabling personalized image generation based on subject images and textual instructions with only forward passes. To address the identity blending problem in the multi-subject generation, FastComposer proposes cross-attention localization supervision during training, enforcing the attention of reference subjects localized to the correct regions in the target images. Naively conditioning on subject embeddings results in subject overfitting. FastComposer proposes delayed subject conditioning in the denoising step to maintain both identity and editability in subject-driven image generation. FastComposer generates images of multiple unseen individuals with different styles, actions, and contexts. It achieves 300$\times$-2500$\times$ speedup compared to fine-tuning-based methods and requires zero extra storage for new subjects. FastComposer paves the way for efficient, personalized, and high-quality multi-subject image creation. Code, model, and dataset are available at https://github.com/mit-han-lab/fastcomposer.

Gang Wang, Zhijie Jerry Shi, Mark Nixon et al.

The fast developing Industrial Internet of Things (IIoT) technologies provide a promising opportunity to build large-scale systems to connect numerous heterogeneous devices into the Internet. Most existing IIoT infrastructures are based on a centralized architecture, which is easier for management but cannot effectively support immutable and verifiable services among multiple parties. Blockchain technology provides many desired features for large-scale IIoT infrastructures, such as decentralization, trustworthiness, trackability, and immutability. This paper presents a blockchain-based IIoT architecture to support immutable and verifiable services. However, when applying blockchain technology to the IIoT infrastructure, the required storage space posts a grant challenge to resource-constrained IIoT infrastructures. To address the storage issue, this paper proposes a hierarchical blockchain storage structure, \textit{ChainSplitter}. Specially, the proposed architecture features a hierarchical storage structure where the majority of the blockchain is stored in the clouds, while the most recent blocks are stored in the overlay network of the individual IIoT networks. The proposed architecture seamlessly binds local IIoT networks, the blockchain overlay network, and the cloud infrastructure together through two connectors, the \textit{blockchain connector} and the \textit{cloud connector}, to construct the hierarchical blockchain storage. The blockchain connector in the overlay network builds blocks in blockchain from data generated in IIoT networks, and the cloud connector resolves the blockchain synchronization issues between the overlay network and the clouds. We also provide a case study to show the efficiency of the proposed hierarchical blockchain storage in a practical Industrial IoT case.

Ioannis Papavasileiou, Wenlong Zhang, Xin Wang et al.

As our population ages, neurological impairments and degeneration of the musculoskeletal system yield gait abnormalities, which can significantly reduce quality of life. Gait rehabilitative therapy has been widely adopted to help patients maximize community participation and living independence. To further improve the precision and efficiency of rehabilitative therapy, more objective methods need to be developed based on sensory data. In this paper, an algorithmic framework is proposed to provide classification of gait disorders caused by two common neurological diseases, stroke and Parkinson's Disease (PD), from ground contact force (GCF) data. An advanced machine learning method, multi-task feature learning (MTFL), is used to jointly train classification models of a subject's gait in three classes, post-stroke, PD and healthy gait. Gait parameters related to mobility, balance, strength and rhythm are used as features for the classification. Out of all the features used, the MTFL models capture the more important ones per disease, which will help provide better objective assessment and therapy progress tracking. To evaluate the proposed methodology we use data from a human participant study, which includes five PD patients, three post-stroke patients, and three healthy subjects. Despite the diversity of abnormalities, the evaluation shows that the proposed approach can successfully distinguish post-stroke and PD gait from healthy gait, as well as post-stroke from PD gait, with Area Under the Curve (AUC) score of at least 0.96. Moreover, the methodology helps select important gait features to better understand the key characteristics that distinguish abnormal gaits and design personalized treatment.