We present a two-stage fine-tuning approach to make the large language model Qwen3 14B "think" natively in Korean. In the first stage, supervised fine-tuning (SFT) on a high-quality Korean reasoning dataset establishes a strong foundation in Korean logical reasoning, yielding notable improvements in Korean-language tasks and even some gains in general reasoning ability. In the second stage, we employ reinforcement learning with a customized Group Relative Policy Optimization (GRPO) algorithm to further enhance both Korean reasoning alignment and overall problem-solving performance. We address critical stability challenges in GRPO training - such as reward hacking and policy collapse - by introducing an oracle judge model that calibrates the reward signal. Our approach achieves stable learning (avoiding the collapse observed in naive GRPO) and leads to steady, incremental performance gains. The final RL-tuned model demonstrates substantially improved results on advanced reasoning benchmarks (particularly math and coding tasks) while maintaining knowledge and language proficiency, successfully conducting its internal chain-of-thought entirely in Korean.

Cloud computing environments are increasingly vulnerable to security threats such as distributed denial-of-service (DDoS) attacks and SQL injection. Traditional security mechanisms, based on rule matching and feature recognition, struggle to adapt to evolving attack strategies. This paper proposes an adaptive security protection framework leveraging deep learning to construct a multi-layered defense architecture. The proposed system is evaluated in a real-world business environment, achieving a detection accuracy of 97.3%, an average response time of 18 ms, and an availability rate of 99.999%. Experimental results demonstrate that the proposed method significantly enhances detection accuracy, response efficiency, and resource utilization, offering a novel and effective approach to cloud computing security.

Programmers spend a significant amount of time reading code during the software development process. This trend is amplified by the emergence of large language models (LLMs) that automatically generate code. However, little is known about the readability of the LLM-generated code and whether it is still important from practitioners' perspectives in this new era. In this paper, we conduct a survey to explore the practitioners' perspectives on code readability in the age of LLMs and investigate the readability of our LLM-based software development agents framework, HULA, by comparing its generated code with human-written code in real-world scenarios. Overall, the findings underscore that (1) readability remains a critical aspect of software development; (2) the readability of our LLM-generated code is comparable to human-written code, fostering the establishment of appropriate trust and driving the broad adoption of our LLM-powered software development platform.

Operations and maintenance (O&M) is a fundamental problem in wind energy systems with far reaching implications for reliability and profitability. Optimizing O&M is a multi-faceted decision optimization problem that requires a careful balancing act across turbine level failure risks, operational revenues, and maintenance crew logistics. The resulting O&M problems are typically solved using large-scale mixed integer programming (MIP) models, which yield computationally challenging problems that require either long-solution times, or heuristics to reach a solution. To address this problem, we introduce a novel decision-making framework for wind farm O&M that builds on a multi-head attention (MHA) models, an emerging artificial intelligence methods that are specifically designed to learn in rich and complex problem settings. The development of proposed MHA framework incorporates a number of modeling innovations that allows explicit embedding of MIP models within an MHA structure. The proposed MHA model (i) significantly reduces the solution time from hours to seconds, (ii) guarantees feasibility of the proposed solutions considering complex constraints that are omnipresent in wind farm O&M, (iii) results in significant solution quality compared to the conventional MIP formulations, and (iv) exhibits significant transfer learning capability across different problem settings.

Efficiently integrating renewable resources into electricity markets is vital for addressing the challenges of matching real-time supply and demand while reducing the significant energy wastage resulting from curtailments. To address this challenge effectively, the incorporation of storage devices can enhance the reliability and efficiency of the grid, improving market liquidity and reducing price volatility. In short-term electricity markets, participants navigate numerous options, each presenting unique challenges and opportunities, underscoring the critical role of the trading strategy in maximizing profits. This study delves into the optimization of day-ahead and balancing market trading, leveraging quantile-based forecasts. Employing three trading approaches with practical constraints, our research enhances forecast assessment, increases trading frequency, and employs flexible timestamp orders. Our findings underscore the profit potential of simultaneous participation in both day-ahead and balancing markets, especially with larger battery storage systems; despite increased costs and narrower profit margins associated with higher-volume trading, the implementation of high-frequency strategies plays a significant role in maximizing profits and addressing market challenges. Finally, we modelled four commercial battery storage systems and evaluated their economic viability through a scenario analysis, with larger batteries showing a shorter return on investment.

Recommendation systems are ubiquitous, from Spotify playlist suggestions to Amazon product suggestions. Nevertheless, depending on the methodology or the dataset, these systems typically fail to capture user preferences and generate general recommendations. Recent advancements in Large Language Models (LLM) offer promising results for analyzing user queries. However, employing these models to capture user preferences and efficiency remains an open question. In this paper, we propose LLMRS, an LLM-based zero-shot recommender system where we employ pre-trained LLM to encode user reviews into a review score and generate user-tailored recommendations. We experimented with LLMRS on a real-world dataset, the Amazon product reviews, for software purchase use cases. The results show that LLMRS outperforms the ranking-based baseline model while successfully capturing meaningful information from product reviews, thereby providing more reliable recommendations.

This research provides a novel framework to estimate the economic, environmental, and social values of electrifying public transit buses, for cities across the world, based on open-source data. Electric buses are a compelling candidate to replace diesel buses for the environmental and social benefits. However, the state-of-art models to evaluate the value of bus electrification are limited in applicability because they require granular and bespoke data on bus operation that can be difficult to procure. Our valuation tool uses General Transit Feed Specification, a standard data format used by transit agencies worldwide, to provide high-level guidance on developing a prioritization strategy for electrifying a bus fleet. We develop physics-informed machine learning models to evaluate the energy consumption, the carbon emissions, the health impacts, and the total cost of ownership for each transit route. We demonstrate the scalability of our tool with a case study of the bus lines in the Greater Boston and Milan metropolitan areas. Detailed Affiliation: U.Vijay, S.Woo, and S.J.Moura are at Department of Civil and Environmental Engineering, University of California-Berkeley, Davis Hall, Berkeley, California, 94720, USA. A.Jain is at Department of Electrical Engineering and Computer Sciences, University of California-Berkeley, Soda Hall, Berkeley, California, 94720, USA. D.Rodriguez and E.Mellekas are at Enel X, North America, Inc., One Marina Park Drive, Boston, 02210, MA, USA. S. Gambacorta is at Enel X, Innovation and Sustainability Global, Smart City, Viale Tor di Quinto, Rome, 00191, Italy. G.Ferrara is at Enel X, Innovation and Sustainability Global, Smart City, Passo Martino, Catania, 95121, Italy. L.Lanuzza is at Enel X, Innovation and Sustainability B2C & B2B Innovation Factory, Viale Tor di Quinto, Rome, 00191, Italy. C.Zulberti and C.Papa are at Enel Foundation, Via Bellini, Rome, 00198, Italy. Vehicle electrification is crucial for reducing the climate impact of the transportation sector, which currently accounts for 16.2% of the global greenhouse gas emissions [22]. Zero-emission electric vehicles can significantly improve the air quality, health, and environmental equity [23], [24].

We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 - 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Machine learning (ML) is an invaluable asset to modern businesses across the board. However, when it comes to ML models, both B2C and B2B companies face the problem of delayed time to market. According to Algorithmia, a vast majority of companies take at least a month or longer to first develop and then deploy their ML model. The reason for this is a complex and often very costly two-stage process.

Undetected machine failures are the most expensive ones. That is why many manufacturing companies are looking for solutions that automate and reduce maintenance costs. Traditional vibrodiagnostic methods can be too late in many cases. Taking readings in the presence of a diagnostician occasionally may not detect a fault in advance. The benefits of predictive maintenance are dependent on the industry or the specific processes that it is applied to. However, Deloitte analyses at that time have already concluded that material cost savings amount to 5 to 10% on average.

This paper proposes a safe reinforcement learning algorithm for generation bidding decisions and unit maintenance scheduling in a competitive electricity market environment. In this problem, each unit aims to find a bidding strategy that maximizes its revenue while concurrently retaining its reliability by scheduling preventive maintenance. The maintenance scheduling provides some safety constraints which should be satisfied at all times. Satisfying the critical safety and reliability constraints while the generation units have an incomplete information of each others' bidding strategy is a challenging problem. Bi-level optimization and reinforcement learning are state of the art approaches for solving this type of problems. However, neither bi-level optimization nor reinforcement learning can handle the challenges of incomplete information and critical safety constraints. To tackle these challenges, we propose the safe deep deterministic policy gradient reinforcement learning algorithm which is based on a combination of reinforcement learning and a predicted safety filter. The case study demonstrates that the proposed approach can achieve a higher profit compared to other state of the art methods while concurrently satisfying the system safety constraints.