We present AROMA (Attentive Reduced Order Model with Attention), a framework designed to enhance the modeling of partial differential equations (PDEs) using local neural fields. Our flexible encoder-decoder architecture can obtain smooth latent representations of spatial physical fields from a variety of data types, including irregular-grid inputs and point clouds. This versatility eliminates the need for patching and allows efficient processing of diverse geometries. The sequential nature of our latent representation can be interpreted spatially and permits the use of a conditional transformer for modeling the temporal dynamics of PDEs. By employing a diffusion-based formulation, we achieve greater stability and enable longer rollouts compared to conventional MSE training. AROMA's superior performance in simulating 1D and 2D equations underscores the efficacy of our approach in capturing complex dynamical behaviors.

As large language models continue to grow in size, parameter-efficient fine-tuning (PEFT) has become increasingly crucial. While low-rank adaptation (LoRA) offers a solution through low-rank updates, its static rank allocation may yield suboptimal results. Adaptive low-rank adaptation (AdaLoRA) improves this with dynamic allocation but remains sensitive to initial and target rank configurations. We introduce AROMA, a framework that automatically constructs layer-specific updates by iteratively building up rank-one components with very few trainable parameters that gradually diminish to zero. Unlike existing methods that employ rank reduction mechanisms, AROMA introduces a dual-loop architecture for rank growth. The inner loop extracts information from each rank-one subspace, while the outer loop determines the number of rank-one subspaces, i.e., the optimal rank. We reset optimizer states to maintain subspace independence. AROMA significantly reduces parameters compared to LoRA and AdaLoRA while achieving superior performance on natural language understanding and commonsense reasoning tasks, offering new insights into adaptive PEFT. The code is available at \href{https://github.com/ShuDun23/AROMA}{AROMA}.

Artificial intelligence can tell Scotch whisky from American whiskey and identify its strongest constituent aromas more reliably than human experts – by using data rather than tasting the drinks. Andreas Grasskamp at the Fraunhofer Institute for Process Engineering and Packaging IVV in Germany and his colleagues trained an AI molecular odour prediction algorithm called OWSum on descriptions of different whiskies. Then, in a study involving 16 samples – nine types of Scotch whisky and seven types of American bourbon or whiskey – they tasked OWSum with telling drinks from the two nations apart based on keyword descriptions of their flavours, such as flowery, fruity, woody or smoky. Using these alone, the AI could tell which country a drink came from with almost 94 per cent accuracy. Because the complex aroma of these spirits is determined by the absence or presence of many chemical compounds, the researchers also fed the AI a reference dataset of 390 molecules commonly found in whiskies.

They arguably have one of the best occupations in the world. But whisky sommeliers may soon have some competition for their jobs – from AI. Scientists have devised machine learning algorithms that can determine whether a whisky is of American or Scotch origin and identify its strongest aromas. And they even outperform human experts, the results show. A whisky's aroma is determined by a complex mixture of odorous compounds, which makes it highly challenging to assess. Panels of human experts are often used to identify the strongest notes of a whisky but these require a significant investment in time, money and training – and agreement between experts is often rare.

Researchers have used the technology to predict the notes that waft off whisky and determine whether a dram was made in the US or Scotland. The work is a step towards automated systems that can predict the complex aroma of whisky from its molecular makeup. Expert panels usually assess woody, smoky, buttery or caramel aromas, which can help to ensure they don't vary substantially between batches of the same product. "The beautiful thing about the AI is that it is very consistent," said Dr Andreas Grasskamp, who led the research at the Fraunhofer Institute for Process Engineering and Packaging in Freising, Germany. "You have this subjectivity still in trained experts. We are not replacing the human nose with this, but we are really supporting it through efficiency and consistency."

We present AROMA (Attentive Reduced Order Model with Attention), a framework designed to enhance the modeling of partial differential equations (PDEs) using local neural fields. Our flexible encoder-decoder architecture can obtain smooth latent representations of spatial physical fields from a variety of data types, including irregular-grid inputs and point clouds. This versatility eliminates the need for patching and allows efficient processing of diverse geometries. The sequential nature of our latent representation can be interpreted spatially and permits the use of a conditional transformer for modeling the temporal dynamics of PDEs. By employing a diffusion-based formulation, we achieve greater stability and enable longer rollouts compared to conventional MSE training. AROMA's superior performance in simulating 1D and 2D equations underscores the efficacy of our approach in capturing complex dynamical behaviors.

The performance of the reward model (RM) is a critical factor in improving the effectiveness of the large language model (LLM) during alignment fine-tuning. There remain two challenges in RM training: 1) training the same RM using various categories of data may cause its generalization performance to suffer from multi-task disturbance, and 2) the human annotation consistency rate is generally only $60\%$ to $75\%$, causing training data to contain a lot of noise. To tackle these two challenges, we introduced the idea of Mixture-of-Experts (MoE) into the field of RM for the first time. We propose the Double-Layer MoE RM (DMoERM). The outer layer MoE is a sparse model. After classifying an input into task categories, we route it to the corresponding inner layer task-specific model. The inner layer MoE is a dense model. We decompose the specific task into multiple capability dimensions and individually fine-tune a LoRA expert on each one. Their outputs are then synthesized by an MLP to compute the final rewards. To minimize costs, we call a public LLM API to obtain the capability preference labels. The validation on manually labeled datasets confirms that our model attains superior consistency with human preference and outstrips advanced generative approaches. Meanwhile, through BoN sampling and RL experiments, we demonstrate that our model outperforms state-of-the-art ensemble methods of RM and mitigates the overoptimization problem. Our code and dataset are available at: https://github.com/quanshr/DMoERM-v1.

Alex Wiltschko opens a black plastic suitcase and pulls out about 60 glass vials. Each contains a different scent. One smells starchy with soft floral notes, like jasmine rice cooking. Another brings to mind ocean air and the white rind of a watermelon. One is like saffron with hints of leather and black tea.

Scent-enriched tours will be accessible to visually impaired people in a way entirely visual exhibitions can never be. As the idea of preserving sensory heritage quietly catches on in the cultural and museum fields, an ambitious project aims to investigate how scents defined communities in the past. ODEUROPA is the first pan-European initiative to use artificial intelligence (AI) to create a library of historic smells. The research team plans to bring some of these aromas from the 17th and 18th century back to life and to preserve them, either by finding words that accurately describe them or by using modern scientific processes to recreate these smells in the lab. "One of our aims is to make cultural experiences more tangible," explained Inger Leemans, professor of cultural history and project lead of ODEUROPA at the Royal Netherlands Academy of Arts and Sciences (KNAW).