搜索结果：Raises

Within the highly successful $Λ$CDM paradigm established with cosmic microwave background (CMB) anisotropy measurements, the optical depth through reionization $τ$ is the most uncertain due both to the difficulty in measuring large-angle polarization and the assumptions made in their interpretation. Currently, for the Planck primary data in the flat $Λ$CDM cosmology with slow-roll inflation and standard reionization, the one-sided 95% upper limit for $τ$ is $τ_{\rm max}=0.0696$. Yet when all current CMB measurements excluding large-angle polarization are combined with baryon acoustic oscillation (BAO) measurements, the one-sided 95% lower limit is an incompatible $τ_{\rm min}=0.074$. If the long-standing low-power feature of the temperature measurements is interpreted as physically originating from inflation then $τ$ inferred from large-angle polarization becomes larger. Marginalizing over templates of the low-power feature based on the generalized slow-roll formalism of inflation raises the Planck maximum to a more compatible $τ_{\rm max}=0.075$ which further increases to $τ_{\rm max} = 0.082$ with the inclusion of all CMB+BAO data. This marginalization does not assess the statist

Retrieval-augmented generation (RAG) systems expose numerous design choices spanning query rewriting, chunking, retrieval depth, reranking, and context compression. In practice, these choices are often configured through heuristics, hindering systematic evaluation and reproducibility across settings. We argue that this challenge is best formulated as RAG architecture search. To support controlled and reproducible study of this problem, we introduce the RAG Intelligence Search Engine (RAISE), a comprehensive framework and benchmark for RAG hyperparameter optimization, which evaluates optimization methods for RAG pipelines under standardized search spaces and budgets. RAISE implements 13 search algorithms and evaluates them across seven public text and multimodal datasets using three random seeds. Our experiments show that optimization performance is highly task-dependent: methods that perform strongly on one dataset may not generalize consistently across others, cautioning against interpreting aggregate rankings as evidence of universally superior strategies. RAISE provides a common experimental substrate for fair, reproducible, and systematic research on RAG hyperparameter optimiza

The basic element of circuit quantum electrodynamics (cQED) is a cavity resonator strongly coupled to a superconducting qubit. Since the inception of the field, the choice of the cavity frequency was, with a few exceptions, been limited to a narrow range around 7 GHz due to a variety of fundamental and practical considerations. Here we report the first cQED implementation, where the qubit remains a regular transmon at about 5 GHz frequency, but the cavity's fundamental mode raises to 21 GHz. We demonstrate that (i) the dispersive shift remains in the conventional MHz range despite the large qubit-cavity detuning, (ii) the quantum efficiency of the qubit readout reaches 8%, (iii) the qubit's energy relaxation quality factor exceeds $10^7$, (iv) the qubit coherence time reproducibly exceeds $100~μ\rm{s}$ and can reach above $300~μ\rm{s}$ with a single echoing $π$-pulse correction. The readout error is currently limited by an accidental resonant excitation of a non-computational state, the elimination of which requires minor adjustments to the device parameters. Nevertheless, we were able to initialize the qubit in a repeated measurement by post-selection with $2\times 10^{-3}$ error

Surface wettability is a critical design parameter for biomedical devices, coatings, and textiles. Contact angle measurements quantify liquid-surface interactions, which depend strongly on liquid formulation. Herein, we present the Robotic Autonomous Imaging Surface Evaluator (RAISE), a closed-loop, self-driving laboratory that is capable of linking liquid formulation optimization with surface wettability assessment. RAISE comprises a full experimental orchestrator with the ability of mixing liquid ingredients to create varying formulation cocktails, transferring droplets of prepared formulations to a high-throughput stage, and using a pick-and-place camera tool for automated droplet image capture. The system also includes an automated image processing pipeline to measure contact angles. This closed loop experiment orchestrator is integrated with a Bayesian Optimization (BO) client, which enables iterative exploration of new formulations based on previous contact angle measurements to meet user-defined objectives. The system operates in a high-throughput manner and can achieve a measurement rate of approximately 1 contact angle measurement per minute. Here we demonstrate RAISE can

As AI systems enter high-stakes domains, evaluation must extend beyond predictive accuracy to include explainability, fairness, robustness, and sustainability. We introduce RAISE (Responsible AI Scoring and Evaluation), a unified framework that quantifies model performance across these four dimensions and aggregates them into a single, holistic Responsibility Score. We evaluated three deep learning models: a Multilayer Perceptron (MLP), a Tabular ResNet, and a Feature Tokenizer Transformer, on structured datasets from finance, healthcare, and socioeconomics. Our findings reveal critical trade-offs: the MLP demonstrated strong sustainability and robustness, the Transformer excelled in explainability and fairness at a very high environmental cost, and the Tabular ResNet offered a balanced profile. These results underscore that no single model dominates across all responsibility criteria, highlighting the necessity of multi-dimensional evaluation for responsible model selection. Our implementation is available at: https://github.com/raise-framework/raise.

By extending the notion of spin of prime ideals, we show that a short character sum conjecture implies that the set of primes raising the level of a certain even Galois representation has density 2/3, as conjectured by Ramakrishna in 1998.

The rapid advancement of generative AI has enabled the creation of highly photorealistic visual content, offering practical substitutes for real images and videos in scenarios where acquiring real data is difficult or expensive. However, reliably substituting real visual content with AI-generated counterparts requires robust assessment of the perceived realness of AI-generated visual content, a challenging task due to its inherent subjective nature. To address this, we conducted a comprehensive human study evaluating the perceptual realness of both real and AI-generated images, resulting in a new dataset, containing images paired with subjective realness scores, introduced as RAISE in this paper. Further, we develop and train multiple models on RAISE to establish baselines for realness prediction. Our experimental results demonstrate that features derived from deep foundation vision models can effectively capture the subjective realness. RAISE thus provides a valuable resource for developing robust, objective models of perceptual realness assessment.

We prove level raising results for $p$-adic automorphic forms on definite unitary groups $U(3)/\mathbb{Q}$ and deduce some intersection points on the eigenvariety. Let $l$ be an inert prime where the level subgroups varies, if there is a non-very-Eisenstein point $φ$ on the old component (generically parametrizing forms old at $l$) satisfying $T_{l}(φ)=l(l^3+1)$, then this point also lies in the new component (generically parametrizing forms new at $l$). This provides a $p$-adic analogue of Bellaïche and Graftieaux's mod $p$ level raising for classical automorphic forms on $U(3)$, and also generalizes James Newton's $p$-adic level raising results for definite quaternion algebras. Key ingredients include abelian Ihara lemma (proved for any definite unitary group $U(n)$) and some duality arguments about certain Hecke modules. Finally we also discuss some methods to construct such points explicitly and further development.

We consider families of reductive complexes related by level-raising operators and originating from an associative algebra. In the main theorem it is shown that the multiple cohomology of that complexes is given by the factor space of products of reduction operators. In particular, we compute explicit torsor structure of the genus $g$ multiple cohomology of the families of horizontal complexes with spaces of of canonical converging reductive differential forms for a $C_2$-cofinite quasiconformal strong-conformal field theory-type vertex operator algebra associated to a complex curve. That provides an equivalence of multiple cohomology to factor spaces of products of sums of reduction functions with actions of the group of local coordinates automorphisms.

When bidders bid on complex objects, they might be unaware of characteristics effecting their valuations. We assume that each buyer's valuation is a sum of independent random variables, one for each characteristic. When a bidder is unaware of a characteristic, he omits the random variable from the sum. We study the seller's decision to raise bidders' awareness of characteristics before a second-price auction with entry fees. Optimal entry fees capture an additional unawareness rent due to unaware bidders misperceiving their probability of winning and the price to be paid upon winning. When raising a bidder's individual awareness of a characteristic with positive expected value, the seller faces a trade-off between positive effects on the expected first order statistic and unawareness rents of remaining unaware bidders on one hand and the loss of the unawareness rent from the newly aware bidder on the other. We present characterization results on raising public awareness together with no versus full information. We discuss the winner's curse due to unawareness of characteristics.

Scientific reasoning requires not only long-chain reasoning processes, but also knowledge of domain-specific terminologies and adaptation to updated findings. To deal with these challenges for scientific reasoning, we introduce RAISE, a step-by-step retrieval-augmented framework which retrieves logically relevant documents from in-the-wild corpus. RAISE is divided into three steps: problem decomposition, logical query generation, and logical retrieval. We observe that RAISE consistently outperforms other baselines on scientific reasoning benchmarks. We analyze that unlike other baselines, RAISE retrieves documents that are not only similar in terms of the domain knowledge, but also documents logically more relevant.

Observations of Starlink V2 Mini satellites during orbit-raising suggest that SpaceX applies brightness mitigation when they reach a height of 357 km. The mean apparent magnitudes for objects below that height threshold is 2.68 while the mean for those above is 6.46. When magnitudes are adjusted to a uniform distance of 1000 km the means are 4.58 and 7.52, respectively. The difference of 2.94 between distance-adjusted magnitudes above and below threshold implies that mitigation is 93% effective in reducing the brightness of orbit-raising spacecraft. Orbit-raising Mini spacecraft have a smaller impact on astronomical observations than higher altitude on-station spacecraft because they are relatively few in number. They also spend less time traversing the sky and spend longer in the Earth's shadow. These low-altitude objects will be more out-of-focus in large telescopes such as the LSST which reduces their impact, too. However, they attract considerable public attention and airline pilots have reported them as Unidentified Aerial Phenomena.

We give an explicit raising operator formula for the modified Macdonald polynomials $\tilde{H}_{μ}(X;q,t)$, which follows from our recent formula for $ abla$ on an LLT polynomial and the Haglund-Haiman-Loehr formula expressing modified Macdonald polynomials as sums of LLT polynomials. Our method just as easily yields a formula for a family of symmetric functions $\tilde{H}^{1,n}(X;q,t)$ that we call $1,n$-Macdonald polynomials, which reduce to a scalar multiple of $\tilde{H}_μ(X;q,t)$ when $n=1$. We conjecture that the coefficients of $1,n$-Macdonald polynomials in terms of Schur functions belong to $\mathbb{N}[q,t]$, generalizing Macdonald positivity.

The distribution of primes raising the level of even Galois representations of tetrahedral type is studied. Data are presented on primes $v\leq 10^8$ raising the level of $3$-adic even representations of various conductors. Based on the data, a conjecture is formulated concerning the distribution of certain lines in the plane. By an application of Wiles' formula, the conjecture is shown to imply that the density of primes raising the level of a $p$-adic even representation is $$\frac{p-1}{p},$$ in agreement with the density of $2/3$ for $p=3$ observed in the data.

Graph-level anomaly detection has become a critical topic in diverse areas, such as financial fraud detection and detecting anomalous activities in social networks. While most research has focused on anomaly detection for visual data such as images, where high detection accuracies have been obtained, existing deep learning approaches for graphs currently show considerably worse performance. This paper raises the bar on graph-level anomaly detection, i.e., the task of detecting abnormal graphs in a set of graphs. By drawing on ideas from self-supervised learning and transformation learning, we present a new deep learning approach that significantly improves existing deep one-class approaches by fixing some of their known problems, including hypersphere collapse and performance flip. Experiments on nine real-world data sets involving nine techniques reveal that our method achieves an average performance improvement of 11.8% AUC compared to the best existing approach.

Raised $k$-Dyck paths are a generalization of $k$-Dyck paths that may both begin and end at a nonzero height. In this paper, we develop closed formulas for the number of raised $k$-Dyck paths from $(0,α)$ to $(\ell,β)$ for all height pairs $α,β\geq 0$, all lengths $\ell \geq 0$, and all $k \geq 2$. We then enumerate raised $k$-Dyck paths with a fixed number of returns to ground, a fixed minimum height, and a fixed maximum height, presenting generating functions (in terms of the generating functions $C_k(t)$ for the $k$-Catalan numbers) when closed formulas aren't tractable. Specializing our results to $k=2$ or to $α< k$ reveal connections with preexisting results concerning height-bounded Dyck paths and "Dyck paths with a negative boundary", respectively.

This work introduces a novel cause-effect relation in Markov decision processes using the probability-raising principle. Initially, sets of states as causes and effects are considered, which is subsequently extended to regular path properties as effects and then as causes. The paper lays the mathematical foundations and analyzes the algorithmic properties of these cause-effect relations. This includes algorithms for checking cause conditions given an effect and deciding the existence of probability-raising causes. As the definition allows for sub-optimal coverage properties, quality measures for causes inspired by concepts of statistical analysis are studied. These include recall, coverage ratio and f-score. The computational complexity for finding optimal causes with respect to these measures is analyzed.