Search

Vision-language models (VLMs) like CLIP have demonstrated impressive zero-shot ability in image classification tasks by aligning text and images but suffer inferior performance compared with task-specific expert models. On the contrary, expert models excel in their specialized domains but lack zero-shot ability for new tasks. How to obtain both the high performance of expert models and zero-shot ability is an important research direction. In this paper, we attempt to demonstrate that by constructing a model hub and aligning models with their functionalities using model labels, new tasks can be solved in a zero-shot manner by effectively selecting and reusing models in the hub. We introduce a novel paradigm, Model Label Learning (MLL), which bridges the gap between models and their functionalities through a Semantic Directed Acyclic Graph (SDAG) and leverages an algorithm, Classification Head Combination Optimization (CHCO), to select capable models for new tasks. Compared with the foundation model paradigm, it is less costly and more scalable, i.e., the zero-shot ability grows with the sizes of the model hub. Experiments on seven real-world datasets validate the effectiveness and efficiency of MLL, demonstrating that expert models can be effectively reused for zero-shot tasks. Our code will be released publicly.

Pre-trained vision-language models like CLIP have shown powerful zero-shot inference ability via image-text matching and prove to be strong few-shot learners in various downstream tasks. However, in real-world scenarios, adapting CLIP to downstream tasks may encounter the following challenges: 1) data may exhibit long-tailed data distributions and might not have abundant samples for all the classes; 2) There might be emerging tasks with new classes that contain no samples at all. To overcome them, we propose a novel framework to achieve efficient and long-tailed generalization, which can be termed as Candle. During the training process, we propose compensating logit-adjusted loss to encourage large margins of prototypes and alleviate imbalance both within the base classes and between the base and new classes. For efficient adaptation, we treat the CLIP model as a black box and leverage the extracted features to obtain visual and textual prototypes for prediction. To make full use of multi-modal information, we also propose cross-modal attention to enrich the features from both modalities. For effective generalization, we introduce virtual prototypes for new classes to make up for their lack of training images. Candle achieves state-of-the-art performance over extensive experiments on 11 diverse datasets while substantially reducing the training time, demonstrating the superiority of our approach. The source code is available at https://github.com/shijxcs/Candle., Comment: Accepted by KDD 2024

In offline Imitation Learning (IL), one of the main challenges is the \textit{covariate shift} between the expert observations and the actual distribution encountered by the agent, because it is difficult to determine what action an agent should take when outside the state distribution of the expert demonstrations. Recently, the model-free solutions introduce the supplementary data and identify the latent expert-similar samples to augment the reliable samples during learning. Model-based solutions build forward dynamic models with conservatism quantification and then generate additional trajectories in the neighborhood of expert demonstrations. However, without reward supervision, these methods are often over-conservative in the out-of-expert-support regions, because only in states close to expert-observed states can there be a preferred action enabling policy optimization. To encourage more exploration on expert-unobserved states, we propose a novel model-based framework, called offline Imitation Learning with Self-paced Reverse Augmentation (SRA). Specifically, we build a reverse dynamic model from the offline demonstrations, which can efficiently generate trajectories leading to the expert-observed states in a self-paced style. Then, we use the subsequent reinforcement learning method to learn from the augmented trajectories and transit from expert-unobserved states to expert-observed states. This framework not only explores the expert-unobserved states but also guides maximizing long-term returns on these states, ultimately enabling generalization beyond the expert data. Empirical results show that our proposal could effectively mitigate the covariate shift and achieve the state-of-the-art performance on the offline imitation learning benchmarks. Project website: \url{https://www.lamda.nju.edu.cn/shaojj/KDD24_SRA/}., Comment: Accepted by KDD2024

Sterile neutrinos that couple to the Standard Model via the neutrino magnetic dipole portals have been extensively studied at various experiments. In this work, we scrutinize these interactions for sterile neutrinos in the mass range of $\unit[0.1]{}-\unit[50]{MeV}$ through the nuclear and electron recoils at various neutrino scattering experiments. For the $e$-flavor specific dipole portal, we demonstrate that Dresden-II can provide leading constraints for $m_N \lesssim \unit[0.5]{MeV}$, setting aside currently unresolved theoretical uncertainties. For the $\mu$-flavor case, we show that the COHERENT experiment can probe a unique parameter region for $m_N$ in the range of $\unit[10]{}-\unit[40]{MeV}$ with the full dataset collected by the CsI[Na] scintillation detector, including both the energy and timing structure of the neutrino beam. We also present limits on the parameter regions of the $\tau$-flavor dipole portal using measurements of the solar neutrino flux from dark matter direct detection experiments., Comment: 8 pages, 2 figures

Large language models (LLMs) have demonstrated impressive performance on reasoning tasks, which can be further improved through few-shot prompting techniques. However, the current evaluation primarily focuses on carefully constructed benchmarks and neglects the consideration of real-world reasoning problems that present missing and contradictory conditions, known as ill-defined problems. Our observations suggest that existing few-shot prompting techniques are ineffective in such scenarios, often providing overconfident answers or hallucination. To further study this problem, we develop a benchmark called Problems with Missing and Contradictory conditions (PMC) and introduce two novel metrics to evaluate the performance of few-shot prompting methods in these scenarios. Our analysis using the PMC benchmark reveals a trade-off dilemma between the performance of mathematical reasoning for well-defined problems and the ability to recognize ill-defined problems. To address the challenges posed by PMC, we propose a novel few-shot prompting method called SMT-LIB Prompting (SLP), which utilizes the SMT-LIB language to model the problems instead of solving them directly. Subsequently, a double-check solving strategy checks the satisfiability and uniqueness of the solution and provides final feedback. Extensive experiments demonstrate the superiority of our SLP approach compared to existing few-shot prompting methods when dealing with problems with missing and contradictory conditions. We will open-source our benchmark and code to facilitate future research., Comment: Preprint. arXiv admin note: text overlap with arXiv:2304.09797

Large language models (LLMs), including both proprietary and open-source models, have showcased remarkable capabilities in addressing a wide range of downstream tasks. Nonetheless, when it comes to practical Chinese legal tasks, these models fail to meet the actual requirements. Proprietary models do not ensure data privacy for sensitive legal cases, while open-source models demonstrate unsatisfactory performance due to their lack of legal knowledge. To address this problem, we introduce LawGPT, the first open-source model specifically designed for Chinese legal applications. LawGPT comprises two key components: legal-oriented pre-training and legal supervised fine-tuning. Specifically, we employ large-scale Chinese legal documents for legal-oriented pre-training to incorporate legal domain knowledge. To further improve the model's performance on downstream legal tasks, we create a knowledge-driven instruction dataset for legal supervised fine-tuning. Our experimental results demonstrate that LawGPT outperforms the open-source LLaMA 7B model. Our code and resources are publicly available at https://github.com/pengxiao-song/LaWGPT and have received 5.7K stars on GitHub., Comment: Technical Report

Vision-language models (VLMs), such as CLIP, have demonstrated impressive zero-shot capabilities for various downstream tasks. Their performance can be further enhanced through few-shot prompt tuning methods. However, current studies evaluate the performance of learned prompts separately on base and new classes. This evaluation lacks practicality for real-world applications since downstream tasks cannot determine whether the data belongs to base or new classes in advance. In this paper, we explore a problem setting called Open-world Prompt Tuning (OPT), which involves tuning prompts on base classes and evaluating on a combination of base and new classes. By introducing Decomposed Prompt Tuning framework (DePT), we theoretically demonstrate that OPT can be solved by incorporating out-of-distribution detection into prompt tuning, thereby enhancing the base-to-new discriminability. Based on DePT, we present a novel prompt tuning approach, namely, Decomposed Context Optimization (DeCoOp), which introduces new-class detectors and sub-classifiers to further enhance the base-class and new-class discriminability. Experimental results on 11 benchmark datasets validate the effectiveness of DePT and demonstrate that DeCoOp outperforms current state-of-the-art methods, providing a significant 2% average accuracy improvement., Comment: Accepted by ICML 2024. Code is available at: https://wnjxyk.github.io/DeCoOp

In this work we discuss the neutrino mass dependent nuclear matrix element (NME) of the neutrinoless double beta decay process and derive the limit on the parameter space of the minimal Type-I seesaw model from the current available experimental data as well as the future sensitivities from the next-generation experiments. Both the explicit many-body calculations and naive extrapolations of the mass dependent NME are employed in the current work. The uncertainties of the theoretical nuclear structure models are taken into account. By combining the latest experimental data from $^{76}$Ge-based experiments, GERDA and MAJORANA, the $^{130}$Te-based experiment, CUORE and the $^{136}$Xe-based experiments, KamLAND-Zen and EXO-200, the bounds on the parameter space of the minimal Type-I seesaw model are obtained and compared with the limits from other experimental probes. Sensitivities for future experiments utilizing $^{76}$Ge-based (LEGEND-1000), $^{82}$Se-based (SuperNEMO), $^{130}$Te based (SNO+II) and $^{136}$Xe-based (nEXO), with a ten-year exposure, are also derived., Comment: 36 pages, 14 figures, the version to be published in JHEP

Atmospheric neutrinos contribute significantly to irreducible backgrounds through their neutral-current (NC) interactions with $^{12}$C nuclei in liquid-scintillator detectors, impacting diffuse supernova neutrino background, nucleon decay, and reactor neutrinos. This paper extends our prior work by systematically studying the NC backgrounds towards the MeV region of reactor neutrinos. We employ contemporary neutrino generator models from GENIE and NuWro for calculations, with a focus on predicting NC background in experimental searches for inverse-beta-decay signals below 100 MeV visible energy. We estimate the systematic uncertainty to our estimation of the NC background using various data-driven neutrino generator models, addressing factors such as the initial neutrino-nucleon NC interaction, the nuclear model, the final-state interaction model, the nucleus deexcitation, and the secondary interaction on final-state particles., Comment: 26 pages, 10 figures

Label quality issues, such as noisy labels and imbalanced class distributions, have negative effects on model performance. Automatic reweighting methods identify problematic samples with label quality issues by recognizing their negative effects on validation samples and assigning lower weights to them. However, these methods fail to achieve satisfactory performance when the validation samples are of low quality. To tackle this, we develop Reweighter, a visual analysis tool for sample reweighting. The reweighting relationships between validation samples and training samples are modeled as a bipartite graph. Based on this graph, a validation sample improvement method is developed to improve the quality of validation samples. Since the automatic improvement may not always be perfect, a co-cluster-based bipartite graph visualization is developed to illustrate the reweighting relationships and support the interactive adjustments to validation samples and reweighting results. The adjustments are converted into the constraints of the validation sample improvement method to further improve validation samples. We demonstrate the effectiveness of Reweighter in improving reweighting results through quantitative evaluation and two case studies., Comment: Submitted to IEEE TVCG

Pulse shape discrimination (PSD) is widely used in particle and nuclear physics. Specifically in liquid scintillator detectors, PSD facilitates the classification of different particle types based on their energy deposition patterns. This technique is particularly valuable for studies of the Diffuse Supernova Neutrino Background (DSNB), nucleon decay, and dark matter searches. This paper presents a detailed investigation of the PSD technique, applied in the DSNB search performed with the Jiangmen Underground Neutrino Observatory (JUNO). Instead of using conventional cut-and-count methods, we employ methods based on Boosted Decision Trees and Neural Networks and compare their capability to distinguish the DSNB signals from the atmospheric neutrino neutral-current background events. The two methods demonstrate comparable performance, resulting in a 50\% to 80\% improvement in signal efficiency compared to a previous study performed for JUNO~\cite{JUNO:2015zny}. Moreover, we study the dependence of the PSD performance on the visible energy and final state composition of the events and find a significant dependence on the presence/absence of $^{11}$C. Finally, we evaluate the impact of the detector effects (photon propagation, PMT dark noise, and waveform reconstruction) on the PSD performance., Comment: 21 pages, 9 figures

Long-tail learning has received significant attention in recent years due to the challenge it poses with extremely imbalanced datasets. In these datasets, only a few classes (known as the head classes) have an adequate number of training samples, while the rest of the classes (known as the tail classes) are infrequent in the training data. Re-sampling is a classical and widely used approach for addressing class imbalance issues. Unfortunately, recent studies claim that re-sampling brings negligible performance improvements in modern long-tail learning tasks. This paper aims to investigate this phenomenon systematically. Our research shows that re-sampling can considerably improve generalization when the training images do not contain semantically irrelevant contexts. In other scenarios, however, it can learn unexpected spurious correlations between irrelevant contexts and target labels. We design experiments on two homogeneous datasets, one containing irrelevant context and the other not, to confirm our findings. To prevent the learning of spurious correlations, we propose a new context shift augmentation module that generates diverse training images for the tail class by maintaining a context bank extracted from the head-class images. Experiments demonstrate that our proposed module can boost the generalization and outperform other approaches, including class-balanced re-sampling, decoupled classifier re-training, and data augmentation methods. The source code is available at https://www.lamda.nju.edu.cn/code_CSA.ashx., Comment: Accepted by NeurIPS 2023

In this work, we introduce the theoretical framework of the phonon-mediated Migdal effect for neutrino-nucleus scattering in semiconductors, considering both the Standard Model and the presence of the neutrino magnetic moment. We calculate the rate of electron excitation resulting from the Migdal effect and observe a substantial coherent enhancement compared to ordinary neutrino-electron scattering. Furthermore, we provide numerical sensitivities for detecting the Migdal effect and constraints on the neutrino magnetic moment, utilizing an experimental setup with reactor neutrinos., Comment: 16 pages, 3 figures, Nucl.Phys.B in press

Contrastive Language-Image Pre-training (CLIP) provides a foundation model by integrating natural language into visual concepts, enabling zero-shot recognition on downstream tasks. It is usually expected that satisfactory overall accuracy can be achieved across numerous domains through well-designed textual prompts. However, we found that their performance in the worst categories is significantly inferior to the overall performance. For example, on ImageNet, there are a total of 10 categories with class-wise accuracy as low as 0\%, even though the overall performance has achieved 64.1\%. This phenomenon reveals the potential risks associated with using CLIP models, particularly in risk-sensitive applications where specific categories hold significant importance. To address this issue, we investigate the alignment between the two modalities in the CLIP model and propose the Class-wise Matching Margin (\cmm) to measure the inference confusion. \cmm\ can effectively identify the worst-performing categories and estimate the potential performance of the candidate prompts. We further query large language models to enrich descriptions of worst-performing categories and build a weighted ensemble to highlight the efficient prompts. Experimental results clearly verify the effectiveness of our proposal, where the accuracy on the worst-10 categories on ImageNet is boosted to 5.2\%, without manual prompt engineering, laborious optimization, or access to labeled validation data.

The fine-tuning paradigm in addressing long-tail learning tasks has sparked significant interest since the emergence of foundation models. Nonetheless, how fine-tuning impacts performance in long-tail learning was not explicitly quantified. In this paper, we disclose that heavy fine-tuning may even lead to non-negligible performance deterioration on tail classes, and lightweight fine-tuning is more effective. The reason is attributed to inconsistent class conditions caused by heavy fine-tuning. With the observation above, we develop a low-complexity and accurate long-tail learning algorithms LIFT with the goal of facilitating fast prediction and compact models by adaptive lightweight fine-tuning. Experiments clearly verify that both the training time and the learned parameters are significantly reduced with more accurate predictive performance compared with state-of-the-art approaches. The implementation code is available at https://github.com/shijxcs/LIFT., Comment: Accepted by ICML 2024

The theoretical framework of the neutrino electron excitation at low energies including the screening effect in semiconductor detectors is developed for the first time, both in the Standard Model and in the presence of the neutrino magnetic moment. We explore the contribution of the screening effect of semiconductors to the neutrino electron excitation based on the linear response theory and calculate the corresponding numerical results. We show that excitation rates from the neutrino magnetic moment are dramatically enhanced {by} the screening effect and the sensitivity can be significantly improved to the level of $10^{-13}\,\mu_{\rm B}$, much better than the current {laboratory and astrophysical} limits., Comment: 7 pages, 2 figures

Geoneutrinos serve as a potent tool for comprehending the radiogenic power and composition of Earth. Although geoneutrinos have been observed in prior experiments, the forthcoming generation of experiments,such as JUNO, will be necessary for fully harnessing their potential. Precise prediction of the crustal contribution is vital for interpreting particlephysics measurements in the context of geo-scientific inquiries. Nonetheless, existing models such as JULOC and GIGJ have limitations in accurately forecasting the crustal contribution. This paper introduces JULOCI, the novel 3-D integrated crustal model of JUNO, which employs seismic, gravity, rock sample, and heat flow data to precisely estimate the geoneutrino signal of the lithosphere. The model indicates elevated concentrations of uranium and thorium in southern China, resulting in unexpectedly strong geoneutrino signals.The accuracy of JULOC-I, coupled with a decade of experimental data, affords JUNO the opportunity to test multiple mantle models. Once operational, JUNO can validate the model predictions and enhance the precision of mantle measurements. All in all, the improved accuracy ofJULOC-I represents a substantial stride towards comprehending the geochemical distribution of the South China crust, offering a valuable tool for investigating the composition and evolution of the Earth through geoneutrinos., Comment: Substantial updates on the model and predictions, submitted version

Multi-label learning has attracted significant attention from both academic and industry field in recent decades. Although existing multi-label learning algorithms achieved good performance in various tasks, they implicitly assume the size of target label space is not huge, which can be restrictive for real-world scenarios. Moreover, it is infeasible to directly adapt them to extremely large label space because of the compute and memory overhead. Therefore, eXtreme Multi-label Learning (XML) is becoming an important task and many effective approaches are proposed. To fully understand XML, we conduct a survey study in this paper. We first clarify a formal definition for XML from the perspective of supervised learning. Then, based on different model architectures and challenges of the problem, we provide a thorough discussion of the advantages and disadvantages of each category of methods. For the benefit of conducting empirical studies, we collect abundant resources regarding XML, including code implementations, and useful tools. Lastly, we propose possible research directions in XML, such as new evaluation metrics, the tail label problem, and weakly supervised XML., Comment: A preliminary version

Semi-supervised learning (SSL) improves model generalization by leveraging massive unlabeled data to augment limited labeled samples. However, currently, popular SSL evaluation protocols are often constrained to computer vision (CV) tasks. In addition, previous work typically trains deep neural networks from scratch, which is time-consuming and environmentally unfriendly. To address the above issues, we construct a Unified SSL Benchmark (USB) for classification by selecting 15 diverse, challenging, and comprehensive tasks from CV, natural language processing (NLP), and audio processing (Audio), on which we systematically evaluate the dominant SSL methods, and also open-source a modular and extensible codebase for fair evaluation of these SSL methods. We further provide the pre-trained versions of the state-of-the-art neural models for CV tasks to make the cost affordable for further tuning. USB enables the evaluation of a single SSL algorithm on more tasks from multiple domains but with less cost. Specifically, on a single NVIDIA V100, only 39 GPU days are required to evaluate FixMatch on 15 tasks in USB while 335 GPU days (279 GPU days on 4 CV datasets except for ImageNet) are needed on 5 CV tasks with TorchSSL., Comment: Accepted by NeurIPS'22 dataset and benchmark track; code at https://github.com/microsoft/Semi-supervised-learning

LAMDA-SSL is open-sourced on GitHub and its detailed usage documentation is available at https://ygzwqzd.github.io/LAMDA-SSL/. This documentation introduces LAMDA-SSL in detail from various aspects and can be divided into four parts. The first part introduces the design idea, features and functions of LAMDA-SSL. The second part shows the usage of LAMDA-SSL by abundant examples in detail. The third part introduces all algorithms implemented by LAMDA-SSL to help users quickly understand and choose SSL algorithms. The fourth part shows the APIs of LAMDA-SSL. This detailed documentation greatly reduces the cost of familiarizing users with LAMDA-SSL toolkit and SSL algorithms.

The detection potential of the active neutrino magnetic moment ($\nu_{a}$MM) and the active sterile transition magnetic moment ($\nu_{s}$MM) of the solar neutrino CE$\nu$NS process in future Dark Matter direct detection experiments is studied and compared with the respective allowed range to explain the Xenon1T excess. We find that the sensitivity to the $\nu_{a}$MM approaches to the level between $10^{-10}\mu_B$ and $10^{-11}\mu_B$, which is dominantly limited by the detection threshold. On the other hand, the future solar neutrino CE$\nu$NS detection would be powerful in probing the $\nu_{s}$MM for the sterile neutrino mass below 10 MeV, which could unambiguously test the $\nu_{s}$MM explanation of the Xenon1T excess. The sensitivity in the general framework with both $\nu_{a}$MM and $\nu_{s}$MM contributions is also derived., Comment: 22 pages, 5 figures

Transit Timing Variation (TTV) of hot Jupiters provides direct observational evidence of planet tidal dissipation. Detecting tidal dissipation through TTV needs high precision transit timings and long timing baselines. In this work, we predict and discuss the potential scientific contribution of SiTian Survey in detecting and analyzing exoplanet TTV. We develop a tidal dissipation detection pipeline for SiTian Survey that aims at time-domain astronomy with 72 1-meter optical telescopes. The pipeline includes the modules of light curve deblending, transit timing obtaining, and TTV modeling. SiTian is capable to detect more than 25,000 exoplanets among which we expect $\sim$50 sources showing evidence of tidal dissipation. We present detection and analysis of tidal dissipating targets, based on simulated SiTian light curves of XO-3b and WASP-161b. The transit light curve modeling gives consistent results within 1$\sigma$ to input values of simulated light curves. Also, the parameter uncertainties predicted by Monte-Carlo Markov Chain are consistent with the distribution obtained from simulating and modeling the light curve 1000 times. The timing precision of SiTian observations is $\sim$ 0.5 minutes with one transit visit. We show that differences between TTV origins, e.g., tidal dissipation, apsidal precession, multiple planets, would be significant, considering the timing precision and baseline. The detection rate of tidal dissipating hot Jupiters would answer a crucial question of whether the planet migrates at an early formation stage or random stages due to perturbations, e.g., planet scattering, secular interaction. SiTian identified targets would be constructive given that the sample would extend tenfold., Comment: Accept for publication to RAA

Semi-supervised learning (SSL) is the branch of machine learning that aims to improve learning performance by leveraging unlabeled data when labels are insufficient. Recently, SSL with deep models has proven to be successful on standard benchmark tasks. However, they are still vulnerable to various robustness threats in real-world applications as these benchmarks provide perfect unlabeled data, while in realistic scenarios, unlabeled data could be corrupted. Many researchers have pointed out that after exploiting corrupted unlabeled data, SSL suffers severe performance degradation problems. Thus, there is an urgent need to develop SSL algorithms that could work robustly with corrupted unlabeled data. To fully understand robust SSL, we conduct a survey study. We first clarify a formal definition of robust SSL from the perspective of machine learning. Then, we classify the robustness threats into three categories: i) distribution corruption, i.e., unlabeled data distribution is mismatched with labeled data; ii) feature corruption, i.e., the features of unlabeled examples are adversarially attacked; and iii) label corruption, i.e., the label distribution of unlabeled data is imbalanced. Under this unified taxonomy, we provide a thorough review and discussion of recent works that focus on these issues. Finally, we propose possible promising directions within robust SSL to provide insights for future research., Comment: We will rewrite this paper

The advent of gadolinium-loaded Super-Kamiokande (SK-Gd) and of the soon-to-start JUNO liquid scintillator detector marks a substantial improvement in the global sensitivity for the Diffuse Supernova Neutrino Background (DSNB). The present article reviews the detector properties most relevant for the DSNB searches in both experiments and estimates the expected signal and background levels. Based on these inputs, we evaluate the sensitivity of both experiments individually and combined. Using a simplified statistical approach, we find that both SK-Gd and JUNO have the potential to reach $>$3$\sigma$ evidence of the DSNB signal within 10 years of measurement. The combined results are likely to enable a $5\sigma$ discovery of the DSNB signal within the next decade., Comment: 21 pages, 6 figures

Dark matter (DM) direct detection experiments are entering the multiple-ton era and will be sensitive to the coherent elastic neutrino nucleus scattering (CE$\nu$NS) of solar neutrinos, enabling the possibility to explore contributions from new physics with light mediators at the low energy range. In this paper we consider light mediator models (scalar, vector and axial vector) and the corresponding contributions to the solar neutrino CE$\nu$NS process. Motivated by the current status of new generation of DM direct detection experiments and the future plan, we study the sensitivity of light mediators in DM direct detection experiments of different nuclear targets and detector techniques. The constraints from the latest $^8$B solar neutrino measurements of XENON-1T are also derived. Finally, We show that the solar neutrino CE$\nu$NS process can provide stringent limitation on the $ L_{\mu}-L_{\tau} $ model with the vector mediator mass below 100 MeV, covering the viable parameter space of the solution to the $ (g-2)_{\mu}$ anomaly., Comment: 25 pages, 5 figures

We discuss the contribution of right-handed neutrinos (RHNs) to the effective neutrino mass of the neutrinoless double beta decay within the minimal type-I seesaw model using the intrinsic seesaw relation of neutrino mass and mixing parameters and the relative mass dependence of the nuclear matrix elements. In the viable parameter space, we find the possibilities of both the enhancement and cancellation to the effective neutrino mass from RHNs. The bounds on the parameter space of the RHNs can be determined with the effective neutrino mass extracted from neutrinoless double beta decay experiments., Comment: 15 pages, 6 figures, PLB published version

Model-independent reactor isotopic cross sections per fission are determined by global fits of the reactor antineutrino data from High-Enriched Uranium (HEU) reactor rates, Low-Enriched Uranium (LEU) reactor rates, and reactor fuel evolution data. Taking account of the implicit quasi-linear relationship between the fission fractions of $^{239}\rm{Pu}$ and $^{241}\rm{Pu}$ in the LEU reactor data, the Inverse-Beta-Decay (IBD) yields and their correlations of the fissionable isotopes $^{235}\rm{U}$, $^{238}\rm{U}$, and Pu's are obtained. The data-driven isotopic IBD yields provide an anomaly-free model for the reactor isotopic cross sections per fission, where better than 1\% accuracy of the expected reactor IBD yields can be achieved for future experiments., Comment: 11 pages, 4 figures. With more discussions and references added, accepted for publication at Physical Review D

We consider neutrino evolution master equations in dense moving and polarized matter consisted of electrons, neutrons, protons and neutrinos. We also take into account the neutrino magnetic moment interaction with a magnetic field. We point out the mechanisms responsible for the neutrino spin precession and provide the expressions for the corresponding interaction Hamiltonians that should be taken into account in theoretical treatments of collective neutrino oscillations., Comment: 4 pages, based on the presentation at the 17th International Conference on Topics in Astroparticle and Underground Physics (TAUP 2021), 26 August - 3 September 2021, hosted by IFIC Valencia and held online

Deep neural networks have been shown to be very powerful methods for many supervised learning tasks. However, they can also easily overfit to training set biases, i.e., label noise and class imbalance. While both learning with noisy labels and class-imbalanced learning have received tremendous attention, existing works mainly focus on one of these two training set biases. To fill the gap, we propose \textit{Prototypical Classifier}, which does not require fitting additional parameters given the embedding network. Unlike conventional classifiers that are biased towards head classes, Prototypical Classifier produces balanced and comparable predictions for all classes even though the training set is class-imbalanced. By leveraging this appealing property, we can easily detect noisy labels by thresholding the confidence scores predicted by Prototypical Classifier, where the threshold is dynamically adjusted through the iteration. A sample reweghting strategy is then applied to mitigate the influence of noisy labels. We test our method on CIFAR-10-LT, CIFAR-100-LT and Webvision datasets, observing that Prototypical Classifier obtains substaintial improvements compared with state of the arts., Comment: 10 pages

While semi-supervised learning (SSL) has received tremendous attentions in many machine learning tasks due to its successful use of unlabeled data, existing SSL algorithms use either all unlabeled examples or the unlabeled examples with a fixed high-confidence prediction during the training progress. However, it is possible that too many correct/wrong pseudo labeled examples are eliminated/selected. In this work we develop a simple yet powerful framework, whose key idea is to select a subset of training examples from the unlabeled data when performing existing SSL methods so that only the unlabeled examples with pseudo labels related to the labeled data will be used to train models. The selection is performed at each updating iteration by only keeping the examples whose losses are smaller than a given threshold that is dynamically adjusted through the iteration. Our proposed approach, Dash, enjoys its adaptivity in terms of unlabeled data selection and its theoretical guarantee. Specifically, we theoretically establish the convergence rate of Dash from the view of non-convex optimization. Finally, we empirically demonstrate the effectiveness of the proposed method in comparison with state-of-the-art over benchmarks., Comment: ICML 2021

The existence of noisy data is prevalent in both the training and testing phases of machine learning systems, which inevitably leads to the degradation of model performance. There have been plenty of works concentrated on learning with in-distribution (IND) noisy labels in the last decade, i.e., some training samples are assigned incorrect labels that do not correspond to their true classes. Nonetheless, in real application scenarios, it is necessary to consider the influence of out-of-distribution (OOD) samples, i.e., samples that do not belong to any known classes, which has not been sufficiently explored yet. To remedy this, we study a new problem setup, namely Learning with Open-world Noisy Data (LOND). The goal of LOND is to simultaneously learn a classifier and an OOD detector from datasets with mixed IND and OOD noise. In this paper, we propose a new graph-based framework, namely Noisy Graph Cleaning (NGC), which collects clean samples by leveraging geometric structure of data and model predictive confidence. Without any additional training effort, NGC can detect and reject the OOD samples based on the learned class prototypes directly in testing phase. We conduct experiments on multiple benchmarks with different types of noise and the results demonstrate the superior performance of our method against state of the arts., Comment: ICCV 2021

Long-tailed learning has attracted much attention recently, with the goal of improving generalisation for tail classes. Most existing works use supervised learning without considering the prevailing noise in the training dataset. To move long-tailed learning towards more realistic scenarios, this work investigates the label noise problem under long-tailed label distribution. We first observe the negative impact of noisy labels on the performance of existing methods, revealing the intrinsic challenges of this problem. As the most commonly used approach to cope with noisy labels in previous literature, we then find that the small-loss trick fails under long-tailed label distribution. The reason is that deep neural networks cannot distinguish correctly-labeled and mislabeled examples on tail classes. To overcome this limitation, we establish a new prototypical noise detection method by designing a distance-based metric that is resistant to label noise. Based on the above findings, we propose a robust framework,~\algo, that realizes noise detection for long-tailed learning, followed by soft pseudo-labeling via both label smoothing and diverse label guessing. Moreover, our framework can naturally leverage semi-supervised learning algorithms to further improve the generalisation. Extensive experiments on benchmark and real-world datasets demonstrate the superiority of our methods over existing baselines. In particular, our method outperforms DivideMix by 3\% in test accuracy. Source code will be released soon.

The supernova neutrino flavor evolution in the presence of the non-trivial neutrino magnetic moment and strong magnetic field is numerically derived using the two-flavor and single-angle approximation. The novel properties of collective neutrino oscillations are studied and distinct patterns of flavor and spin-flavor spectral splits are presented. Finally we also discuss how the neutrino magnetic moment affects the observable supernova neutrino energy spectra., Comment: 10 pages, 4 figures

Future large liquid-scintillator (LS) detectors are competitive with and complementary to the water-Cherenkov detectors on the searches for diffuse supernova neutrino background and nucleon decay. In a companion paper, we have performed a systematic calculation of the neutral-current (NC) background induced by atmospheric neutrino interactions on $^{12}{\rm C}$ nuclei in LS detectors, which are expected to be crucially important for the experimental searches for the diffuse supernova neutrino background and nucleon decay. In this paper, we perform a systematic study on the measurement of the NC background and evaluate the associated uncertainties. We first exploit the characteristics of the NC background, in particular, the multiplicities of neutrons and pions, and the possible association with unstable residual nuclei. It turns out that the neutron multiplicity distribution is very powerful to discriminate among different models. Then, we develop a maximum-likelihood method to allow an {\it in situ} measurement of the NC interactions with a triple-coincidence signature. Finally, a data-driven approach is proposed to evaluate the uncertainty of the NC background in the search for the diffuse supernova neutrino background. We conclude that future large LS experiments like JUNO (Jiangmen Underground Neutrino Observatory) will be able to make a unique contribution to the worldwide data set to improve the prediction of atmospheric neutrino NC interactions on $^{12}$C., Comment: 20 pages, 7 figures, more discussions, version to appear in PRD

The experimental searches for diffuse supernova neutrino background and proton decay in next-generation large liquid-scintillator (LS) detectors are competitive with and complementary to those in the water-Cherenkov detectors. In this paper, we carry out a systematic study of the dominant background induced by atmospheric neutrinos via their neutral-current (NC) interactions with the $^{12}{\rm C}$ nuclei in the LS detectors. The atmospheric neutrino fluxes at the location of Jiangmen Underground Neutrino Observatory (JUNO) are used, as the JUNO detector is obviously a suitable representative for future LS detectors. Then, we implement the sophisticated generators \texttt{GENIE} and \texttt{NuWro} to simulate the neutrino interactions with the carbon nuclei, and the package \texttt{TALYS} to deal with the deexcitations of final-state nuclei. Finally, the event rates for the production of additional nucleons, $\gamma$'s, $\alpha$'s, pions and kaons are obtained and categorized, and the systematic uncertainty of the NC background represented by a variety of data-driven nuclear models is estimated. The implications of the NC background from atmospheric neutrinos for the detection of diffuse supernova neutrino background and proton decay are also discussed., Comment: 30 pages, 10 figures, more discussions, version to appear in PRD

Before massive stars heavier than $(8 \cdots 10)$ solar masses evolve to the phase of a gravitational core collapse, they will emit a huge number of MeV-energy neutrinos that are mainly produced in the thermal processes and nuclear weak interactions. The detection of such pre-supernova (pre-SN) neutrinos could provide an important and independent early warning for the optical observations of core-collapse SNe. In this paper, we investigate the prospects of future large liquid-scintillator detectors for the observation of pre-SN neutrinos in both $\bar{\nu}^{}_e + p \to e^+ + n$ and $\nu (\bar{\nu}) + e^- \to \nu (\bar{\nu}) + e^-$ reaction channels, where $\nu$ ($\bar{\nu}$) denotes neutrinos (antineutrinos) of all three flavors. We propose a quantitative assessment of the capability in terms of three working criteria, namely, how far the SN distance can be covered, how long the early warning before the core collapse can be sent out, and how well the direction pointing to the SN can be determined. The dependence of the final results on the different models of progenitor stars, neutrino flavor conversions and the relevant backgrounds is also discussed., Comment: 26 pages, 8 figures and 3 tables, more discussions added, to be published in JCAP