Abstract--Generative AI poses both opportunities and risks for solving inverse design problems in the sciences. Generative tools provide the ability to expand and refine a search space autonomously, but do so at the cost of exploring low-quality regions until sufficiently fine tuned. Here, we propose a queue prioritization algorithm that combines generative modeling and active learning in the context of a distributed workflow for exploring complex design spaces. We find that incorporating an active learning model to prioritize top design candidates can prevent a generative AI workflow from expending resources on nonsensical candidates and halt potential generative model decay. For an existing generative AI workflow for discovering novel molecular structure candidates for carbon capture, our active learning approach significantly increases the number of high-quality candidates identified by the generative model. We find that, out of 1000 novel candidates, our workflow without active learning can generate an average of 281 high-performing candidates, while our proposed prioritization with active learning can generate an average 604 high-performing candidates. In recent years, Generative AI (GenAI) models have significantly changed how computational screening workflows are designed and executed.

When you purchase through links in our articles, we may earn a small commission. TCL's latest top of-the-line smart lock adds a doorbell and better battery to its impressive hands-free security tech. The TCL D2 Pro Palm Vein Smart Lock looks and operates like a device you'd see in a sci-fi movie. The touchless unlock feature is both captivating and rock-solid, but you might be disappointed that it doesn't support Apple Home or Matter. First, we ditched our housekeys for a numerical code that unlocked our doors, then we moved up to sometimes balky fingerprint readers.

We Asked Audio Pros to Blind Test Headphones. When it comes to choosing a good pair of headphones, what happens when you take features, design, and brand awareness out of the equation, and leave it all up to the sound? All products featured on WIRED are independently selected by our editors. However, we may receive compensation from retailers and/or from purchases of products through these links. What makes a really great pair of headphones? The basic answer used to be sound quality, but modern headphones offer so much more than just audio chops.

Anthropic is starting to train its models on new Claude chats. If you're using the bot and don't want your chats used as training data, here's how to opt out. Anthropic is prepared to repurpose conversations users have with its Claude chatbot as training data for its large language models--unless those users opt out. Previously, the company did not train its generative AI models on user chats. When Anthropic's privacy policy updates on October 8 to start allowing for this, users will have to opt out, or else their new chat logs and coding tasks will be used to train future Anthropic models. "All large language models, like Claude, are trained using large amounts of data," reads part of Anthropic's blog explaining why the company made this policy change.

We consider the challenging practical problem of optimizing the power production of a complex of hydroelectric power plants, which involves control over three continuous action variables, uncertainty in the amount of water inflows and a variety of constraints that need to be satisfied. We propose a policy-search-based approach coupled with predictive modelling to address this problem. This approach has some key advantages compared to other alternatives, such as dynamic programming: the policy representation and search algorithm can conveniently incorporate domain knowledge; the resulting policies are easy to interpret, and the algorithm is naturally parallelizable. Our algorithm obtains a policy which outperforms the solution found by dynamic programming both quantitatively and qualitatively.

In this paper, we investigate three cross-facility data streaming architectures, Direct Streaming (DTS), Proxied Streaming (PRS), and Managed Service Streaming (MSS). We examine their architectural variations in data flow paths and deployment feasibility, and detail their implementation using the Data Streaming to HPC (DS2HPC) architectural framework and the SciStream memory-to-memory streaming toolkit on the production-grade Advanced Computing Ecosystem (ACE) infrastructure at Oak Ridge Leadership Computing Facility (OLCF). We present a workflow-specific evaluation of these architectures using three synthetic workloads derived from the streaming characteristics of scientific workflows. Through simulated experiments, we measure streaming throughput, round-trip time, and overhead under work sharing, work sharing with feedback, and broadcast and gather messaging patterns commonly found in AI-HPC communication motifs. Our study shows that DTS offers a minimal-hop path, resulting in higher throughput and lower latency, whereas MSS provides greater deployment feasibility and scalability across multiple users but incurs significant overhead. PRS lies in between, offering a scalable architecture whose performance matches DTS in most cases.

The evaluation of time series forecasting models is hindered by a critical lack of high-quality benchmarks, leading to a potential illusion of progress. Existing datasets suffer from issues ranging from pre-training data contamination in the age of LLMs to the causal and description leakage prevalent in early multimodal designs. To address this, we formalize the core principles of high-fidelity benchmarking, focusing on data sourcing integrity, strict causal soundness, and structural clarity. We introduce Fidel-TS, a new large-scale benchmark built from the ground up on these principles by sourcing data from live APIs. Our extensive experiments validate this approach by exposing the critical biases and design limitations of prior benchmarks. Furthermore, we conclusively demonstrate that the causal relevance of textual information is the key factor in unlocking genuine performance gains in multimodal forecasting.

Abstract--Dynamic V oltage and Frequency Scaling (DVFS) is essential for enhancing energy efficiency in mobile platforms. However, traditional heuristic-based governors are increasingly inadequate for managing the complexity of heterogeneous System-on-Chip designs and diverse application workloads. Although reinforcement learning approaches offer improved performance, their poor generalization capability and reliance on extensive retraining for each hardware and application combination leads to significant deployment costs. In this work, we observe that device and application metadata inherently encapsulate valuable knowledge for DVFS, presenting an opportunity to overcome these limitations. We formulate DVFS for heterogeneous devices and applications as a multi-task reinforcement learning problem. We introduce MetaDVFS, which is a metadata-guided framework that systematically leverages metadata to discover and transfer shared knowledge across DVFS tasks. Evaluations on five Google Pixel devices running six applications show that MetaDVFS achieves up to 17% improvement in Performance-Power Ratio and up to 26% improvement in Quality of Experience. Compared to state-of-the-art methods, MetaDVFS delivers 70.8% faster adaptation (3.5 1.1 vs. 11.8 5.2 minutes) and 5.8-27.6% These results establish MetaDVFS as an effective and scalable solution for DVFS deployment in heterogeneous mobile environments. Dynamic V oltage and Frequency Scaling (DVFS) is an essential technique for effectively improving energy efficiency in battery-powered mobile platforms. DVFS adjusts the operating voltage and frequency of a device in response to current workload demands [1]. Experimental evaluations report energy savings exceeding 26% on mobile MPSoCs where DVFS functions compared to statically managed systems [2]. Traditional DVFS policies typically rely on heuristic-based governors, such as ondemand and schedutil, which make frequency decisions based primarily on simple utilization metrics. Jinqi Y an, Qianlong Sang, Yili Gong, Chuang Hu, and Dazhao Cheng are with the School of Computer Science, Wuhan University.