Vision Transformers (ViTs) achieve outstanding performance by effectively capturing long-range dependencies between image patches (tokens). However, the high computational cost and memory requirements of ViTs present challenges for model compression and deployment on edge devices. In this study, we introduce a new framework, Sparse Structure Exploration and Re-optimization (SERo), specifically designed to maximize pruning efficiency in ViTs. Our approach focuses on (1) hardware-friendly pruning that fully compresses pruned parameters instead of zeroing them out, (2) separating the exploration and re-optimization phases \red{in order to find the optimal structure among various possible sparse structures}, and (3) using a simple gradient magnitude-based criterion for pruning a pre-trained model. SERo iteratively refines pruning masks to identify optimal sparse structures and then re-optimizes the pruned structure, reducing computational costs while maintaining model performance. Experimental results indicate that SERo surpasses existing pruning methods across various ViT models in both performance and computational efficiency. For example, SERo achieves a 69% reduction in computational cost and a 2.4x increase in processing speed for DeiT-Base model, with only a 1.55% drop in accuracy. Implementation code: https://github.com/Ahnho/SERo/

Compute Express Link (CXL) memory sharing, persistent memory, and other related technologies allow data to survive crash events. A key challenge is ensuring that data is consistent after crashes such that it can be safely accessed. While there has been much work on bug-finding tools for persistent memory programs, these tools cannot guarantee that a program is crash-consistent. In this paper, we present a language, CrashLang , and its type system, that together guarantee that well-typed data structure implementations written in CrashLang are crash-consistent. CrashLang leverages the well-known commit-store pattern in which a single store logically commits an entire data structure operation. In this paper, we prove that well-typed CrashLang programs are crash-consistent, and provide a prototype implementation of the CrashLang compiler. We have evaluated CrashLang on five benchmarks: the Harris linked list, the Treiber stack, the Michael–Scott queue, a Read-Copy-Update binary search tree, and a Cache-Line Hash Table. We experimentally verified that each implementation correctly survives crashes.

Domain adaptation for object detection typically entails transferring knowledge from one visible domain to another visible domain. However, there are limited studies on adapting from the visible to the thermal domain, because the domain gap between the visible and thermal domains is much larger than expected, and traditional domain adaptation can not successfully facilitate learning in this situation. To overcome this challenge, we propose a Distinctive Dual-Domain Teacher (D3T) framework that employs distinct training paradigms for each domain. Specifically, we segregate the source and target training sets for building dual-teachers and successively deploy exponential moving average to the student model to individual teachers of each domain. The framework further incorporates a zigzag learning method between dual teachers, facilitating a gradual transition from the visible to thermal domains during training. We validate the superiority of our method through newly designed experimental protocols with wellknown thermal datasets, i.e., FLIR and KAIST. Source code is available at https://github.com/EdwardDo69/D3T.