Abstract Background Dyslipidemia is a key factor causing cardio cerebrovascular diseases, and the total cholesterol (TC) is an important lipid indicator among them. Studies have shown that environmental factors have a strong association with TC levels. Previous studies only focused on the seasonal variation of TC level and the short-term effects of some environmental factors on TC level over time, and few studies explored the geographical distribution of TC level and quantified the impact of environmental factors in space. Methods Based on blood test data which was from China Health and Retirement Longitudinal Study (Charls) database, this study selected the TC level test data of middle-aged and elderly people in China in 2011 and 2015, and collected data from 665 meteorological stations and 1496 air pollutant monitoring stations in China. After pretreatment, the spatial distribution map of TC level was prepared and the regional statistics were made. GeoDetector and geographically weighted regression (GWR) were used to measure the relationship between environmental factors and TC level. Results The TC level of middle-aged and elderly in China was higher in females than in males, and higher in urban areas than in rural areas, showing a clustered distribution. The high values were mainly in South China, Southwest China and North China. Temperature, humidity, PM 10 and PM 2.5 were significant environmental factors affecting TC level of middle-aged and elderly people. The impact of pollutants was more severe in northern China, and TC level in southern China was mainly affected by meteorological factors. Conclusions There were gender and urban-rural differences in TC levels among the middle-aged and elderly population in China, showing aggregation in geographical distribution. Meteorological factors and air pollutants may be very important control factors, and their influencing mechanism needs further study.
Generative Pre-trained Transformers (GPTs) have demonstrated remarkable performance across diverse domains through the extensive scaling of model parameters. Recent works observe the redundancy across the transformer blocks and develop compression methods by structured pruning of the unimportant blocks. However, such straightforward elimination will always provide irreversible performance degradation. In this paper, we propose FuseGPT, a novel methodology to recycle the pruned transformer blocks to further recover the model performance. Firstly we introduce a new importance detection metric, Macro Influence (MI), to detect the long-term influence of each transformer block by calculating their loss of information after removal. Then we propose group-level layers fusion, which adopts the parameters in layers of the unimportant blocks and injects them into the corresponding layers inside the neighboring blocks. The fusion is not one-off but through iterative parameter updates by lightweight group-level fine-tuning. Specifically, these injected parameters are frozen but weighted with learnable rank decomposition matrices to reduce the overhead during fine-tuning. Our approach not only works well on large language models but also on large multimodal models. The experiments have shown that, by using modest amounts of data, FuseGPT can outperform previous works in both perplexity and zero-shot task performance.
Single-image super-resolution (SISR) is an important task in image processing, which aims to enhance the resolution of imaging systems. Recently, SISR has made a huge leap and has achieved promising results with the help of deep learning (DL). In this survey, we give an overview of DL-based SISR methods and group them according to their targets, such as reconstruction efficiency, reconstruction accuracy, and perceptual accuracy. Specifically, we first introduce the problem definition, research background, and the significance of SISR. Secondly, we introduce some related works, including benchmark datasets, upsampling methods, optimization objectives, and image quality assessment methods. Thirdly, we provide a detailed investigation of SISR and give some domain-specific applications of it. Fourthly, we present the reconstruction results of some classic SISR methods to intuitively know their performance. Finally, we discuss some issues that still exist in SISR and summarize some new trends and future directions. This is an exhaustive survey of SISR, which can help researchers better understand SISR and inspire more exciting research in this field. An investigation project for SISR is provided in https://github.com/CV-JunchengLi/SISR-Survey.
Quantization is a critical technique employed across various research fields for compressing deep neural networks (DNNs) to facilitate deployment within resource-limited environments. This process necessitates a delicate balance between model size and performance. In this work, we explore knowledge distillation (KD) as a promising approach for improving quantization performance by transferring knowledge from high-precision networks to low-precision counterparts. We specifically investigate feature-level information loss during distillation and emphasize the importance of feature-level network quantization perception. We propose a novel quantization method that combines feature-level distillation and contrastive learning to extract and preserve more valuable information during the quantization process. Furthermore, we utilize the hyperbolic tangent function to estimate gradients with respect to the rounding function, which smoothens the training procedure. Our extensive experimental results demonstrate that the proposed approach achieves competitive model performance with the quantized network compared to its full-precision counterpart, thus validating its efficacy and potential for real-world applications.
Recent years have seen rising research in logic synthesis recipe generation to improve the Quality-of-Result (QoR). However, existing approaches typically have low efficiency and are stuck at local optima. In this work, we propose a logic synthesis optimization framework, AlphaSyn, that incorporates a domain-specific Monte Carlo tree search (MCTS) algorithm. AlphaSyn enables exploration across the entire search space while optimizing sampling points utilization. We further develop a synthesis-specific upper confidence bound for trees (SynUCT) algorithm for the selection phase and a well-designed learning strategy to enhance the stability of the MCTS algorithm. The AlphaSyn algorithm is fully parallelized for efficiency with asynchronous MCTS exploration and significance-base resource allocation. For standard-cell technology mapping on the ASAP 7nm library among other tasks, experimental results show that AlphaSyn outperforms SOTA FlowTune with an average 8.74% area reduction and $\boldsymbol{1.24}\times$ runtime speedup.
Al2O3-supported Cu and Ni monometallic as well as Cu–Ni bimetallic catalysts were synthesized using a coprecipitation method and studied for the in-situ hydrogenation of furfural (FAL) with isopropanol as the solvent and hydrogen donor. The Cu–Ni bimetallic catalysts showed improved activity toward the production of 2-methylfuran (2-MF) and 2-methyltetrahydrofuran (2-MTHF) over that of monometallic catalysts. The results indicated that isopropanol exhibited better performance than methanol for the in-situ hydrogenation of FAL to produce 2-MF and 2-MTHF under the same conditions. The reaction conditions such as the copper–nickel ratios, catalyst loading amount, reaction temperature, and time were optimized. After the reaction was complete, the supported Cu–Ni bimetallic catalyst could be reused four times without a significant loss in catalytic activity.
Large language models (LLMs) achieve impressive performance by scaling model parameters, but this comes with significant inference overhead. Feed-forward networks (FFNs), which dominate LLM parameters, exhibit high activation sparsity in hidden neurons. To exploit this, researchers have proposed using a mixture-of-experts (MoE) architecture, where only a subset of parameters is activated. However, existing approaches often require extensive training data and resources, limiting their practicality. We propose CMoE (Carved MoE), a novel framework to efficiently carve MoE models from dense models. CMoE achieves remarkable performance through efficient expert grouping and lightweight adaptation. First, neurons are grouped into shared and routed experts based on activation rates. Next, we construct a routing mechanism without training from scratch, incorporating a differentiable routing process and load balancing. Using modest data, CMoE produces a well-designed, usable MoE from a 7B dense model within five minutes. With lightweight fine-tuning, it achieves high-performance recovery in under an hour. We make our code publicly available at https://github.com/JarvisPei/CMoE.
The ever-growing complexity of modern VLSI circuits brings about a substantial increase in the design cycle. As for logic synthesis, how to efficiently obtain physical characteristics of a design for subsequent design space exploration emerges as a critical issue. In this paper, we propose ${\mathsf{LSTP}}$, an ML-based logic synthesis predictor, which can rapidly predict the post-synthesis timing of a broad range of circuit designs. Specifically, we explicitly take optimization sequences into consideration so that we can comprehend the synergy between optimization passes and their effects on netlists. Experimental results demonstrate that we outperform state-of-the-art remarkably.
'/%3e%3cuse xlink:href='%23a' transform='rotate(144 450 300)'/%3e%3cuse xlink:href='%23a' transform='rotate(216 450 300)'/%3e%3cuse xlink:href='%23a' transform='rotate(288 450 300)'/%3e%3c/svg%3e)
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)