搜索结果：Luna

Creating photorealistic, animatable 3D human avatars from monocular images still largely depends on Linear Blend Skinning (LBS) and parametric body models, which constrain expressivity and often introduce artifacts due to imperfect fitting. We propose LUNA, an LBS-free universal neural animation model that directly maps multiple 2D controls like images, keypoints, sketches, and unseen characters into 3D Gaussian deformations, bypassing explicit body fitting. At its core, a transformer-based motion regressor disentangles global rigid motion from fine-grained local dynamics to capture both coherent movement and subtle non-rigid effects. To resolve the inherent ambiguity of 2D-to-3D lifting while scaling beyond fitted datasets, we introduce hybrid supervision that distills soft structural priors from an LBS teacher and a loss that supports training on both limited fitted data and large in-the-wild unlabeled videos. Extensive experiments show LUNA achieves competitive visual fidelity compared to LBS-based approaches, while delivering realistic human motion and zero-shot cross-identity generalization across diverse driving modalities. To the best of our knowledge, LUNA is the first end-

While large language models (LLMs) offer promising reasoning capabilities, their integration into safety-critical driving systems is hindered by limited reasoning diversity, high computational overhead, and static learning paradigms. To address these challenges, we propose LUNA-AD, a lightweight uncertainty-aware language model with lifelong learning for autonomous driving (AD). LUNA-AD features a tri-system architecture that reconciles complex multimodal behavioral reasoning, efficient deployment, and continual refinement. We design a multi-agent analytical system to generate uncertainty-aware decision-making demonstrations through diverse hypothesis exploration. A dual-head lightweight heuristic model is distilled to unify the inference of decision distributions and textual explanations while enabling efficient deployment. Furthermore, a reflection-driven lifelong learning mechanism operates on multimodal decision outputs and preserves strategic diversity, allowing for the refinement of candidate decisions and rationales via closed-loop feedback to enhance driving robustness. Extensive experiments on nuPlan benchmarks demonstrate that LUNA-AD achieves state-of-the-art success rat

Scaling attention faces a critical bottleneck: the $\mathcal{O}(n^2)$ quadratic computational cost of softmax attention, which limits its application in long-sequence domains. While linear attention mechanisms reduce this cost to $\mathcal{O}(n)$, they typically rely on fixed random feature maps, such as random Fourier features or hand-crafted functions. This reliance on static, data-agnostic kernels creates a fundamental trade-off, forcing practitioners to sacrifice significant model accuracy for computational efficiency. We introduce \textsc{LUNA}, a kernelized linear attention mechanism that eliminates this trade-off, retaining linear cost while matching and surpassing the accuracy of quadratic attention. \textsc{LUNA} is built on the key insight that the kernel feature map itself should be learned rather than fixed a priori. By parameterizing the kernel, \textsc{LUNA} learns a feature basis tailored to the specific data and task, overcoming the expressive limitations of fixed-feature methods. \textsc{Luna} implements this with a learnable feature map that induces a positive-definite kernel and admits a streaming form, yielding linear time and memory scaling in the sequence leng

The $^{12}$C+$^{12}$C fusion reaction plays a crucial role in stellar evolution, including the occurrence of supernova explosions, and in the synthesis of the chemical elements. However, our understanding of its cross section remains severely deficient, particularly below $E_\textrm{cm}=2.5$\,MeV, the energy range of interest for astrophysics. To address these unresolved issues, the LUNA collaboration will conduct a dedicated study of the $^{12}$C+$^{12}$C reaction at the Bellotti Ion Beam Facility (Bellotti IBF) located deep underground within the Gran Sasso National Laboratory (LNGS) in Italy. Based on the combination of passive and active shields, this campaign aims to achieve unprecedented sensitivity in measuring the cross sections of the two key reaction channels, $^{12}$C($^{12}$C,$α$)$^{20}$Ne and $^{12}$C($^{12}$C,$p$)$^{23}$Na in the low-energy regime via $γ$-ray detection. Here, we report on a sensitivity study for the upcoming campaign with a focus on the characterization of two detectors, namely a HPGe detector and a NaI(Tl) array. Furthermore, their intrinsic contamination is thoroughly investigated since this could potentially influence the overall sensitivity. Assum

Nuclear reactions are responsible for the chemical evolution of stars, galaxies and the Universe. Unfortunately, at temperatures of interest for nuclear astrophysics, the cross-sections of the thermonuclear reactions are in the pico-femto-barn range and thus measuring them in the laboratory is extremely challenging. In this framework, major steps forward were made with the advent of underground nuclear astrophysics, pioneered by the Laboratory for Underground Nuclear Astrophysics (LUNA). The cosmic background reduction by several orders of magnitude obtained at LUNA, however, needs to be combined with high-performance detectors and dedicated shieldings to obtain the required sensitivity. In the present paper, we report on the recent and future detector-shielding designs at LUNA.

Electroencephalography (EEG) offers a non-invasive lens into human brain activity, but building large-scale models is hampered by topological heterogeneity: each public EEG data defines its own electrode layout, limiting generalization. We introduce LUNA (Latent Unified Network Architecture), a self-supervised foundation model that reconciles disparate electrode geometries while scaling linearly -- not quadratically -- with channel count. LUNA compresses multi-channel EEG into a fixed-size, topology-agnostic latent space via learned queries and cross-attention. Downstream transformer blocks then operate exclusively on this latent representation using patch-wise temporal self-attention, decoupling computation from electrode count. Pre-trained on TUEG and Siena (over 21,000 hours of raw EEG across diverse montages) using a masked-patch reconstruction objective, LUNA transfers effectively to four downstream tasks: abnormality detection, artifact rejection, slowing classification, and emotion recognition. It demonstrates highly competitive performance across several benchmarks, achieving state-of-the-art results on TUAR and TUSL, e.g., 0.921 AUROC on TUAR, while reducing FLOPs by 300x

A precise determination of the proton capture rates on oxygen is mandatory to predict the abundance ratios of oxygen isotopes in a stellar environment where the hydrogen burning is active. The 17O(p,γ)18F reaction, in particular, plays a crucial role in AGB nucleosynthesis as well as in explosive hydrogen burning occurring in novae and type I supernovae. At the temperature of interest for the former scenario (from 20 MK to 80 MK) the main contribution to the astrophysical reaction rate comes from the Er = 64.5 keV resonance. The strength of this resonance is presently determined only through indirect measurements, with an adopted value ωγ = 16(3) peV. A new high sensitivity setup has been installed at LUNA, located at Laboratori Nazionali del Gran Sasso. The underground location of LUNA 400kV guarantees a reduction of the cosmic ray background by several orders of magnitude. The residual background was further reduced by a dedicated shielding. On the other hand the 4πBGO detector efficiency was optimized installing an aluminum target chamber and holder. With about 400 C accumulated on Ta2O5 targets, highly enriched in 17O, the LUNA collaboration has performed the first ever direct

The NeNa-MgAl cycles are involved in the synthesis of Ne, Na, Mg, and Al isotopes. The $^{20}$Ne($p,γ$)$^{21}$Na (Q = 2431.68 keV) reaction is the first and slowest reaction of the NeNa cycle and it controls the speed at which the entire cycle proceeds. At the state of the art, the uncertainty on the 20Ne(p,γ)21Na reaction rate affects the production of the elements in the NeNa cycle. In particular, in the temperature range from 0.1 GK to 1 GK, the rate is dominated by the 366 keV resonance corresponding to the excited state of EX = 2797.5 keV and by the direct capture component. The present study focus on the study of the 366 keV resonance and the direct capture below 400 keV. At LUNA (Laboratory for Underground Nuclear Astrophysics) the $^{20}$Ne($p,γ$)$^{21}$Na reaction has been measured using the intense proton beam delivered by the LUNA 400 kV accelerator and a windowless differential-pumping gas target. The products of the reaction are detected with two high-purity germanium detectors. The experimental details and preliminary results on the 366 keV resonance and on the direct capture component at very low energies will be shown, together with their possible impact on the $^{2

The ${}^{\mathsf{12}}\mathsf{C}(\mathsf{p},γ)$ reaction cross section is currently under investigation in the low-background environment of the Laboratory for Underground Nuclear Astrophysics (LUNA). It is being studied using different types of solid targets, and employing two complementary detection techniques: HPGe spectroscopy and activation counting. To reduce systematic uncertainties, targets have been accurately characterized and their degradation under the intense beam of the LUNA-400 accelerator monitored. We present the experimental techniques and the corresponding analyses used to extract the reaction cross section.

Real-time guardrails require evaluation that is accurate, cheap, and fast - yet today's default, LLM-as-a-judge (LLMAJ), is slow, expensive, and operationally non-deterministic due to multi-token generation. We present Luna-2, a novel architecture that leverages decoder-only small language models (SLMs) into a deterministic evaluation model to reliably compute complex task-specific LLMAJ metrics (e.g. toxicity, hallucination, tool selection quality, etc.) at an accuracy at par or higher than LLMAJ using frontier LLMs while drastically reducing the cost and latency of computation. Each metric is implemented as a lightweight LoRA/PEFT head on top of a shared SLM backbone, enabling hundreds of specialized metrics to run concurrently on a single GPU, deployable locally next to AI systems in a privacy-preserving and latency optimizing manner. Across content safety and hallucination benchmarks, Luna-2 matches the accuracy of state-of-the-art LLM-based evaluators while reducing inference cost by over 80x and latency by over 20x. In this paper, we outline the model architecture, training methodology and report real-world empirical results on accuracy, latency, and throughput results. In pr

Studies of charged-particle reactions for low-energy nuclear astrophysics require high sensitivity, which can be achieved by means of detection setups with high efficiency and low backgrounds, to obtain precise measurements in the energy region of interest for stellar scenarios. High-efficiency total absorption spectroscopy is an established and powerful tool for studying radiative capture reactions, particularly if combined with the cosmic background reduction by several orders of magnitude obtained at the Laboratory for Underground Nuclear Astrophysics (LUNA). We present recent improvements in the detection setup with the Bismuth Germanium Oxide (BGO) detector at LUNA, aiming to reduce high-energy backgrounds and to increase the summing detection efficiency. The new design results in enhanced sensitivity of the BGO setup, as we demonstrate and discuss in the context of the first direct measurement of the 65 keV resonance ($E_{x} = 5672$ keV) of the $^{17}$O($p,γ$)$^{18}$F reaction. Moreover, we show two applications of the BGO detector, which exploit its segmentation. In case of complex $γ$-ray cascades, e.g. the de-excitation of $E_{x} = 5672$ keV in $^{18}$F, the BGO segmentati

Qubit readout is a critical operation in quantum computing systems, which maps the analog response of qubits into discrete classical states. Deep neural networks (DNNs) have recently emerged as a promising solution to improve readout accuracy . Prior hardware implementations of DNN-based readout are resource-intensive and suffer from high inference latency, limiting their practical use in low-latency decoding and quantum error correction (QEC) loops. This paper proposes LUNA, a fast and efficient superconducting qubit readout accelerator that combines low-cost integrator-based preprocessing with Look-Up Table (LUT) based neural networks for classification. The architecture uses simple integrators for dimensionality reduction with minimal hardware overhead, and employs LogicNets (DNNs synthesized into LUT logic) to drastically reduce resource usage while enabling ultra-low-latency inference. We integrate this with a differential evolution based exploration and optimization framework to identify high-quality design points. Our results show up to a 10.95x reduction in area and 30% lower latency with little to no loss in fidelity compared to the state-of-the-art. LUNA enables scalable,

The evaluation of Natural Language Generation (NLG) models has gained increased attention, urging the development of metrics that evaluate various aspects of generated text. LUNA addresses this challenge by introducing a unified interface for 20 NLG evaluation metrics. These metrics are categorized based on their reference-dependence and the type of text representation they employ, from string-based n-gram overlap to the utilization of static embeddings and pre-trained language models. The straightforward design of LUNA allows for easy extension with novel metrics, requiring just a few lines of code. LUNA offers a user-friendly tool for evaluating generated texts.

In early May 2022, the Terra ecosystem collapsed after the algorithmic stablecoin failed to maintain its peg. Emergency measures were taken by Terraform Labs (TFL) in an attempt to protect Luna and UST, but then were abruptly abandoned by TFL for Luna 2.0 several days later. At this time, the Luna Classic blockchain has been left crippled and in limbo for the last two months. In the face of impossible odds, the Luna Classic community has self organized and rallied to build and restore the blockchain. This technical document outlines the steps we, the community, have taken towards the emergency management of the Luna Classic blockchain in the weeks after the UST depeg. We outline precisely what would be implemented on-chain to mitigate the concerns of affected stakeholders, and build trust for external partners, exchanges, and third-party developers. For the Luna Classic community, validators, and developers, this outlines concrete steps on how passed governance can and will be achieved. We openly audit our own code and welcome any feedback for improvement. Let us move forward together as the true community blockchain.

Retriever Augmented Generation (RAG) systems have become pivotal in enhancing the capabilities of language models by incorporating external knowledge retrieval mechanisms. However, a significant challenge in deploying these systems in industry applications is the detection and mitigation of hallucinations: instances where the model generates information that is not grounded in the retrieved context. Addressing this issue is crucial for ensuring the reliability and accuracy of responses generated by large language models (LLMs) in diverse industry settings. Current hallucination detection techniques fail to deliver accuracy, low latency, and low cost simultaneously. We introduce Luna: a DeBERTA-large (440M) encoder, finetuned for hallucination detection in RAG settings. We demonstrate that Luna outperforms GPT-3.5 and commercial evaluation frameworks on the hallucination detection task, with 97% and 91% reduction in cost and latency, respectively. Luna is lightweight and generalizes across multiple industry verticals and out-of-domain data, making it an ideal candidate for industry LLM applications.

We qualitatively examine the accuracy and fidelity between two diffusion-based image generation systems, namely DALL-E 2 and Luna, which have massive differences in training datasets, algorithmic approaches, prompt resolvement, and output upscaling. The methodology used is a qualitative benchmark created by Saharia et al. and in our research we conclude that DALL-E 2 significantly edges Luna in both alignment and fidelity comparisons.

Transformers are widely used in NLP tasks. However, current approaches to leveraging transformers to understand language expose one weak spot: Number understanding. In some scenarios, numbers frequently occur, especially in semi-structured data like tables. But current approaches to rich-number tasks with transformer-based language models abandon or lose some of the numeracy information - e.g., breaking numbers into sub-word tokens - which leads to many number-related errors. In this paper, we propose the LUNA framework which improves the numerical reasoning and calculation capabilities of transformer-based language models. With the number plugin of NumTok and NumBed, LUNA represents each number as a whole to model input. With number pre-training, including regression loss and model distillation, LUNA bridges the gap between number and vocabulary embeddings. To the best of our knowledge, this is the first work that explicitly injects numeracy capability into language models using Number Plugins. Besides evaluating toy models on toy tasks, we evaluate LUNA on three large-scale transformer models (RoBERTa, BERT, TabBERT) over three different downstream tasks (TATQA, TabFact, CrediTra

The quadratic computational and memory complexities of the Transformer's attention mechanism have limited its scalability for modeling long sequences. In this paper, we propose Luna, a linear unified nested attention mechanism that approximates softmax attention with two nested linear attention functions, yielding only linear (as opposed to quadratic) time and space complexity. Specifically, with the first attention function, Luna packs the input sequence into a sequence of fixed length. Then, the packed sequence is unpacked using the second attention function. As compared to a more traditional attention mechanism, Luna introduces an additional sequence with a fixed length as input and an additional corresponding output, which allows Luna to perform attention operation linearly, while also storing adequate contextual information. We perform extensive evaluations on three benchmarks of sequence modeling tasks: long-context sequence modeling, neural machine translation and masked language modeling for large-scale pretraining. Competitive or even better experimental results demonstrate both the effectiveness and efficiency of Luna compared to a variety