Large Language Models (LLMs) have shown remarkable capability in knowledge-intensive scenarios, such as electronic design automation (EDA) tool documentation question answering (QA), due to their ability to process and generate contextually rich, domain-specific information. Evaluating LLM outputs is paramount, as it directly impacts their accuracy, effectiveness, and trustworthiness in practical applications. In this paper, we introduce MAEDA, a novel LLM-powered multi-agent evaluation framework that utilizes multiple fine-tuned LLM agents working collaboratively to assess common error types encountered in EDA tool documentation QA. Specifically, we design customized point-to-point alignment and chain-of-thought (CoT) reasoning strategies tailored to specific agents, enhancing both fine-tuning and inference capabilities. Experimental results demonstrate that MAEDA outperforms state-of-the-art (SOTA) general-purpose and cross-domain evaluation frameworks in accurately identifying error types specific to this domain. Our benchmark is publicly available at https://github.com/Rayzzz14/MAEDA-DATE26/.

Effective mapping of dataflow graphs onto Coarse-Grained Reconfigurable Arrays necessitates compiler-architecture co-design, yet existing approaches frequently assume single-cycle operations despite real-world applications often involving multi-cycle operations that constrain achievable clock frequencies. To address this, we propose FLAME, a novel framework supporting three execution strategies (exclusive, distributed, inclusive) specifically designed for multi-cycle operations, with co-designed compiler and hardware support. Our evaluations demonstrate that FLAME not only surpasses prior methods in performance and but also enables flexible exploration of these operations. The framework achieves average speedups of 2.21× over baseline CGRA and 1.49× over prior state-of-the-art framework while highlighting the distinct characteristics of each strategy.

Large language models (LLMs) have revolutionized natural language processing (NLP) domain by achieving state-of-the-art performance across a range of benchmarks. However, nonlinear operations in LLMs significantly contribute to inference latency and present unique challenges that have not been encountered previously. Addressing these challenges requires accelerators that combine efficiency, flexibility, and support for user-defined precision. Our analysis reveals that Coarse-Grained Reconfigurable Arrays (CGRAs) provide an effective solution, offering a balance of performance and flexibility tailored to domain-specific workloads.

Large language models (LLMs) have revolutionized natural language processing (NLP) tasks by achieving state-of-the-art performance across a range of benchmarks. Central to the success of these models is the integration of sophisticated architectural components aimed at improving training stability, convergence speed, and generalization capabilities. Among these components, normalization operation, such as layer normalization (LayerNorm), emerges as a pivotal technique, offering substantial benefits to the overall model performance. However, previous studies have indicated that normalization operations can substantially elevate processing latency and energy usage. In this work, we adopt the principles of algorithm and hardware co-design, introducing a holistic normalization accelerating method named HAAN. The evaluation results demonstrate that HAAN can achieve significantly better hardware performance compared to state-of-the-art solutions.

The complexity of EDA tools necessitates the development of advanced documentation query answering systems to enhance user efficiency and reduce the associated learning curve. Recent innovations in the use of Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) for EDA tool documentation have demonstrated significant progress; however, these approaches typically lack the multi-modal capabilities required to effectively handle visual data, such as circuit layout and GUI screenshots provided through user input. To address the concern, we introduce a multi-modal RAG system that incorporates two domain-customized modules: a multi-modal retriever model finetuned by the customized bilevel hard negative mining (BHNM) strategy, and a vision large language model (VLLM) finetuned using a tailored extract-score-answer pipeline. Moreover, we have manually curated ORD-MMBench, a multi-modal QA benchmark comprising 120 high-quality question-document-answer triplets based on OpenROAD documentation. Experimental results demonstrate that our customized RAG framework outperforms state-of-the-art multi-modal RAG flows and models on ORD-MMBench.

To improve the coordinated scheduling between the sending and receiving ends of large-scale renewable energy bases (LREBs) in desert regions, this paper proposes a day-ahead multi-objective optimization method. The approach considers both the power output characteristics of the base and the load demand at the receiving end. A bi-level optimization model is constructed, aiming to minimize the standard deviation of residual load at the receiving-end grid and maximize the utilization of renewable energy. The model jointly optimizes thermal unit commitment and power transmission profiles. Simulation results demonstrate that the proposed method achieves a better trade-off between economic performance and renewable integration. Compared with single-objective strategies, it effectively alleviates peak-shaving pressure at the receiving end and reduces renewable energy curtailment at the sending end. The findings confirm the effectiveness of the proposed method for day-ahead scheduling in high-penetration renewable systems and provide theoretical support for flexible transmission and coordinated dispatch in desert LREBs.

This article proposes$\textsf {IncreMacro}$, a novel approach for macro placement refinement in the context of integrated circuit (IC) design. The suggested approach iteratively and incrementally optimizes the placement of macros in order to enhance IC layout routability and timing performance. To achieve this,$\textsf {IncreMacro}$utilizes several methods, including kd-tree-based macro diagnosis, gradient-based macro shifting, constraint-graph-based LP for macro legalization, and diffusion-based cell migration. By employing these techniques iteratively,$\textsf {IncreMacro}$meets two critical solution requirements of macro placement: 1) pushing macros toward the chip boundary and 2) preserving the original macro relative positional relationship. The proposed approach has been incorporated into$\textsf {AutoDMP}$and$\textsf {DREAMPlace}~4.0$, and is evaluated on seven RISC-V benchmark circuits and four TILOS macro placement circuit designs at the 7-nm technology node. Experimental results show that, compared with the macro placement solution provided by$\textsf {AutoDMP}~(\textsf {DREAMPlace}~4.0$), our approach reduces routed wirelength by 15.1% (14.9%), improves the routed worst negative slack (WNS) and total negative slack (TNS) by 99.9 (82.6%) and 99.9% (81.3%), and reduces the total power consumption by 4.4% (4.3%). Meanwhile, compared with$\textsf {IncreMacro}$[1], our approach augmented with the cell migration algorithm improves the routed WNS and TNS by 24.7% and 23.1%, and remains the average routed wirelength and total power consumption almost unchanged.

With the advent of the information age, the function and value of information begin to be fully understood and used by people. As a comprehensive sector of the national economy, banks need to use advanced management system to improve their service ability and service efficiency if they want to maintain their competitiveness and expand their profitability in the fiercely competitive society. This design system uses cross-platform Qt Creator as the system development tool, builds the interface of banking system on Qt Creator, adopts C/S architecture, uses Mysql database to store the managed information and the data are added, deleted, modified and checked. The system is mainly composed of teller interface and client interface. The Banking business mainly includes the functions of opening an account, depositing, withdrawing, transferring, inquiring the balance, etc., which realizes the basic business management of the bank and can accurately and timely display the required information to customers.

In order to solve the problem of poor infrastructure and lack of automatic disaster prevention and mitigation measures in response to frost disaster in Northern Shaanxi mountain Apple planting area, resulting in production reduction or even no harvest, a frost disaster prevention and mitigation system suitable for mountainous areas was designed and implemented. The system is based on LoRa (low-power long-range) wireless communication technology, which is composed of meteorological monitoring unit, gateway, server and defrosting equipment control unit. Through collecting and analyzing the spatial and temporal distribution characteristics of ground temperature in mountainous orchards in frost weather, the system provides early warning of frost disaster and automatically controls the start and stop of defrosting equipment. The system is tested under the condition of simulating the temporal and spatial distribution of ground temperature when frost comes. The system can meet the requirements of automatic control of defrosting equipment.

This paper discusses the intuitive meanings of critical resource and gives its strict definition for the purpose of reducing project makespan. And then, an algorithm to identify critical resources is presented and the complexity of the proposed algorithm is analyzed. The results of a computational example manifest that we can shorten project makespan, decrease idle resources and improve the efficiency of resources utilization by increasing critical resources in some intervals.

This paper proposes the concepts of node's early (later) realization time (ER, LR), node's early (later) possible realization time (EPR, LPR), and gives the formulas of ER, LR, EPR, LPR, total float. And then, we classify the preconditions which determine the change of activities' total floats. Furthermore, the laws of the influence of a change in an activity's duration on its total float are explored and a computational example is presented.

This paper analyzes the shortcomings of activity critical index (CAT), significance index (SI), cruciality index (CRT) and sensitive analysis. And then activity critical comprehensive index (ACCI) is proposed. Computational examples manifest that ACCI offsets the shortcomings of CAI, SI, CRI and sensitive analysis in some extent.

Criticality analysis of activity and path is an important problem in resource-constrained projects. This paper proposes the method of minimal feasible sets (MFS) for the identification of critical activities and critical sequences. And then, we compare MFS with seven related previous methods. Computational results manifest that MFS does not perform inferior to other methods at least and excel them in most situations in establishing resource links and computing floats.

This paper proposes a classification for the relationships between non-critical activities. And then, we explore the influence of a change in a non-critical activity's duration to its early (later) realization time, and give sensitivity analysis of a non-critical activity's early (later) realization time to other activities' total floats. The results of a computational example manifest the efficiency and effectiveness of the proposed methods.