arXiv subjects

Yu-Feng Li

Publications and source records attributed to Yu-Feng Li.

At least 73 records · Page 4Linked to original sources

arXiv2025

Enabling Small Models for Zero-Shot Selection and Reuse through Model Label Learning

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.

cs.AI↗

arXiv2024

Robust Semi-Supervised Learning in Open Environments

Semi-supervised learning (SSL) aims to improve performance by exploiting unlabeled data when labels are scarce. Conventional SSL studies typically assume close environments where important factors (e.g., label, feature, distribution) between labeled and unlabeled data are consistent. However, more practical tasks involve open environments where important factors between labeled and unlabeled data are inconsistent. It has been reported that exploiting inconsistent unlabeled data causes severe performance degradation, even worse than the simple supervised learning baseline. Manually verifying the quality of unlabeled data is not desirable, therefore, it is important to study robust SSL with inconsistent unlabeled data in open environments. This paper briefly introduces some advances in this line of research, focusing on techniques concerning label, feature, and data distribution inconsistency in SSL, and presents the evaluation benchmarks. Open research problems are also discussed for reference purposes.

cs.LG↗

arXiv2024

You Only Submit One Image to Find the Most Suitable Generative Model

Deep generative models have achieved promising results in image generation, and various generative model hubs, e.g., Hugging Face and Civitai, have been developed that enable model developers to upload models and users to download models. However, these model hubs lack advanced model management and identification mechanisms, resulting in users only searching for models through text matching, download sorting, etc., making it difficult to efficiently find the model that best meets user requirements. In this paper, we propose a novel setting called Generative Model Identification (GMI), which aims to enable the user to identify the most appropriate generative model(s) for the user's requirements from a large number of candidate models efficiently. To our best knowledge, it has not been studied yet. In this paper, we introduce a comprehensive solution consisting of three pivotal modules: a weighted Reduced Kernel Mean Embedding (RKME) framework for capturing the generated image distribution and the relationship between images and prompts, a pre-trained vision-language model aimed at addressing dimensionality challenges, and an image interrogator designed to tackle cross-modality issues. Extensive empirical results demonstrate the proposal is both efficient and effective. For example, users only need to submit a single example image to describe their requirements, and the model platform can achieve an average top-4 identification accuracy of more than 80%.

cs.CV↗

arXiv2024

Fully Test-time Adaptation for Tabular Data

Tabular data plays a vital role in various real-world scenarios and finds extensive applications. Although recent deep tabular models have shown remarkable success, they still struggle to handle data distribution shifts, leading to performance degradation when testing distributions change. To remedy this, a robust tabular model must adapt to generalize to unknown distributions during testing. In this paper, we investigate the problem of fully test-time adaptation (FTTA) for tabular data, where the model is adapted using only the testing data. We identify three key challenges: the existence of label and covariate distribution shifts, the lack of effective data augmentation, and the sensitivity of adaptation, which render existing FTTA methods ineffective for tabular data. To this end, we propose the Fully Test-time Adaptation for Tabular data, namely FTAT, which enables FTTA methods to robustly optimize the label distribution of predictions, adapt to shifted covariate distributions, and suit a variety of tasks and models effectively. We conduct comprehensive experiments on six benchmark datasets, which are evaluated using three metrics. The experimental results demonstrate that FTAT outperforms state-of-the-art methods by a margin.

cs.LG↗

arXiv2024

New calculation of the geo-neutrino energy spectrum and its implication

The energy spectrum of geo-neutrinos plays a vital role in the experimental measurement of geo-neutrinos that have profound implications for both particle physics and earth sciences. In this letter, we present a state-of-the-art calculation of the energy spectrum of geo-neutrinos originating from the beta decay of Uranium-238 and Thorium-232. Our calculation is underpinned by the latest updates in the nuclear database, accounts for previously overlooked forbidden transitions, and incorporates advanced corrections for the beta decay. This brand new geo-neutrino flux model, compared to the widely-used estimates from Enomoto, reveals notable distinction in the energy spectrum shape because of our comprehensive approach. When considering the inverse beta decay (IBD) detection process, our findings show a significant deviation in the predicted IBD yield of around 4% for Uranium-238 and 9% for Thorium-232 decay chains. The implications of using the new geo-neutrino flux model for the experimental analysis are substantial, potentially affecting the analysis results of geo-neutrino measurements of KamLAND and Borexino by around 10% to 20%. Our study represents a significant advancement in geo-neutrino research, establishing a new benchmark for accuracy and reliability in the field.

hep-ph↗

arXiv2024

Neuro-Symbolic Data Generation for Math Reasoning

A critical question about Large Language Models (LLMs) is whether their apparent deficiency in mathematical reasoning is inherent, or merely a result of insufficient exposure to high-quality mathematical data. To explore this, we developed an automated method for generating high-quality, supervised mathematical datasets. The method carefully mutates existing math problems, ensuring both diversity and validity of the newly generated problems. This is achieved by a neuro-symbolic data generation framework combining the intuitive informalization strengths of LLMs, and the precise symbolic reasoning of math solvers along with projected Markov chain Monte Carlo sampling in the highly-irregular symbolic space. Empirical experiments demonstrate the high quality of data generated by the proposed method, and that the LLMs, specifically LLaMA-2 and Mistral, when realigned with the generated data, surpass their state-of-the-art counterparts.

cs.AI↗

arXiv2024

Learning for Long-Horizon Planning via Neuro-Symbolic Abductive Imitation

Recent learning-to-imitation methods have shown promising results in planning via imitating within the observation-action space. However, their ability in open environments remains constrained, particularly in long-horizon tasks. In contrast, traditional symbolic planning excels in long-horizon tasks through logical reasoning over human-defined symbolic spaces but struggles to handle observations beyond symbolic states, such as high-dimensional visual inputs encountered in real-world scenarios. In this work, we draw inspiration from abductive learning and introduce a novel framework \textbf{AB}ductive \textbf{I}mitation \textbf{L}earning (ABIL) that integrates the benefits of data-driven learning and symbolic-based reasoning, enabling long-horizon planning. Specifically, we employ abductive reasoning to understand the demonstrations in symbolic space and design the principles of sequential consistency to resolve the conflicts between perception and reasoning. ABIL generates predicate candidates to facilitate the perception from raw observations to symbolic space without laborious predicate annotations, providing a groundwork for symbolic planning. With the symbolic understanding, we further develop a policy ensemble whose base policies are built with different logical objectives and managed through symbolic reasoning. Experiments show that our proposal successfully understands the observations with the task-relevant symbolics to assist the imitation learning. Importantly, ABIL demonstrates significantly improved data efficiency and generalization across various long-horizon tasks, highlighting it as a promising solution for long-horizon planning. Project website: \url{https://www.lamda.nju.edu.cn/shaojj/KDD25_ABIL/}.

cs.LG↗

arXiv2024

Neutrino Electromagnetic Properties

Neutrinos are neutral in the Standard Model, but they have tiny charge radii generated by radiative corrections. In theories Beyond the Standard Model, neutrinos can also have magnetic and electric moments and small electric charges (millicharges). We review the general theory of neutrino electromagnetic form factors, which reduce, for ultrarelativistic neutrinos and small momentum transfers, to the neutrino charges, effective charge radii, and effective magnetic moments. We discuss the phenomenology of these electromagnetic neutrino properties and we review the existing experimental bounds. We briefly review also the electromagnetic processes of astrophysical neutrinos and the neutrino magnetic moment portal in the presence of sterile neutrinos.

hep-ph↗

arXiv2024

Vision-Language Models are Strong Noisy Label Detectors

Recent research on fine-tuning vision-language models has demonstrated impressive performance in various downstream tasks. However, the challenge of obtaining accurately labeled data in real-world applications poses a significant obstacle during the fine-tuning process. To address this challenge, this paper presents a Denoising Fine-Tuning framework, called DeFT, for adapting vision-language models. DeFT utilizes the robust alignment of textual and visual features pre-trained on millions of auxiliary image-text pairs to sieve out noisy labels. The proposed framework establishes a noisy label detector by learning positive and negative textual prompts for each class. The positive prompt seeks to reveal distinctive features of the class, while the negative prompt serves as a learnable threshold for separating clean and noisy samples. We employ parameter-efficient fine-tuning for the adaptation of a pre-trained visual encoder to promote its alignment with the learned textual prompts. As a general framework, DeFT can seamlessly fine-tune many pre-trained models to downstream tasks by utilizing carefully selected clean samples. Experimental results on seven synthetic and real-world noisy datasets validate the effectiveness of DeFT in both noisy label detection and image classification.

cs.LG↗

arXiv2024

Neutrinoless double beta decay in the minimal type-I seesaw model: mass-dependent nuclear matrix element, current limits and future sensitivities

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.

hep-ph↗

arXiv2024

Migdal effect of Phonon-mediated neutrino nucleus scattering in semiconductor detectors

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.

hep-ph↗

arXiv2024

Offline Imitation Learning with Model-based Reverse Augmentation

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/}.

cs.LG↗

arXiv2024

Efficient and Long-Tailed Generalization for Pre-trained Vision-Language Model

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.

cs.CV↗

arXiv2024

Neutrino magnetic dipole portal with low energy neutrino nucleus scattering data

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 $μ$-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 $τ$-flavor dipole portal using measurements of the solar neutrino flux from dark matter direct detection experiments.

hep-ph↗

arXiv2024

LawGPT: A Chinese Legal Knowledge-Enhanced Large Language Model

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.

cs.CL↗

arXiv2024

Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts

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.

cs.CV↗

arXiv2024

DeCoOp: Robust Prompt Tuning with Out-of-Distribution Detection

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.

cs.CV↗

arXiv2024

Pulse shape discrimination technique for diffuse supernova neutrino background search with JUNO

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.

hep-ex↗