arXiv subjects

Peng Zhang

Publications and source records attributed to Peng Zhang.

At least 397 records · Page 22Linked to original sources

arXiv2024

Application of Deep Learning Methods for Distinguishing Gamma-Ray Bursts from Fermi/GBM TTE Data

To investigate GRBs in depth, it is crucial to develop an effective method for identifying GRBs accurately. Current criteria, e.g., onboard blind search, ground blind search, and target search, are limited by manually set thresholds and perhaps miss GRBs, especially for sub-threshold events. We propose a novel approach that utilizes convolutional neural networks (CNNs) to distinguish GRBs and non-GRBs directly. We structured three CNN models, plain-CNN, ResNet, and ResNet-CBAM, and endeavored to exercise fusing strategy models. Count maps of NaI detectors onboard Fermi/GBM were employed as the input samples of datasets and models were implemented to evaluate their performance on different time scale data. The ResNet-CBAM model trained on 64 ms dataset achieves high accuracy overall, which includes residual and attention mechanism modules. The visualization methods of Grad-CAM and t-SNE explicitly displayed that the optimal model focuses on the key features of GRBs precisely. The model was applied to analyze one-year data, accurately identifying approximately 98% of GRBs listed in the Fermi burst catalog, 8 out of 9 sub-threshold GRBs, and 5 GRBs triggered by other satellites, which demonstrated the deep learning methods could effectively distinguish GRBs from observational data. Besides, thousands of unknown candidates were retrieved and compared with the bursts of SGR J1935+2154 for instance, which exemplified the potential scientific value of these candidates indeed. Detailed studies on integrating our model into real-time analysis pipelines thus may improve their accuracy of inspection, and provide valuable guidance for rapid follow-up observations of multi-band telescopes.

astro-ph.HE↗

arXiv2024

Detecting Neutrinos from Supernova Bursts in PandaX-4T

Neutrinos from core-collapse supernovae are essential for the understanding of neutrino physics and stellar evolution. The dual-phase xenon dark matter detectors can provide a way to track explosions of galactic supernovae by detecting neutrinos through coherent elastic neutrino-nucleus scatterings. In this study, a variation of progenitor masses as well as explosion models are assumed to predict the neutrino fluxes and spectra, which result in the number of expected neutrino events ranging from 6.6 to 13.7 at a distance of 10 kpc over a 10-second duration with negligible backgrounds at PandaX-4T. Two specialized triggering alarms for monitoring supernova burst neutrinos are built. The efficiency of detecting supernova explosions at various distances in the Milky Way is estimated. These alarms will be implemented in the real-time supernova monitoring system at PandaX-4T in the near future, providing the astronomical communities with supernova early warnings.

hep-ex↗

arXiv2024

MuseChat: A Conversational Music Recommendation System for Videos

Music recommendation for videos attracts growing interest in multi-modal research. However, existing systems focus primarily on content compatibility, often ignoring the users' preferences. Their inability to interact with users for further refinements or to provide explanations leads to a less satisfying experience. We address these issues with MuseChat, a first-of-its-kind dialogue-based recommendation system that personalizes music suggestions for videos. Our system consists of two key functionalities with associated modules: recommendation and reasoning. The recommendation module takes a video along with optional information including previous suggested music and user's preference as inputs and retrieves an appropriate music matching the context. The reasoning module, equipped with the power of Large Language Model (Vicuna-7B) and extended to multi-modal inputs, is able to provide reasonable explanation for the recommended music. To evaluate the effectiveness of MuseChat, we build a large-scale dataset, conversational music recommendation for videos, that simulates a two-turn interaction between a user and a recommender based on accurate music track information. Experiment results show that MuseChat achieves significant improvements over existing video-based music retrieval methods as well as offers strong interpretability and interactability.

cs.LG↗

arXiv2024

Learning Expressive And Generalizable Motion Features For Face Forgery Detection

Previous face forgery detection methods mainly focus on appearance features, which may be easily attacked by sophisticated manipulation. Considering the majority of current face manipulation methods generate fake faces based on a single frame, which do not take frame consistency and coordination into consideration, artifacts on frame sequences are more effective for face forgery detection. However, current sequence-based face forgery detection methods use general video classification networks directly, which discard the special and discriminative motion information for face manipulation detection. To this end, we propose an effective sequence-based forgery detection framework based on an existing video classification method. To make the motion features more expressive for manipulation detection, we propose an alternative motion consistency block instead of the original motion features module. To make the learned features more generalizable, we propose an auxiliary anomaly detection block. With these two specially designed improvements, we make a general video classification network achieve promising results on three popular face forgery datasets.

cs.CV↗

arXiv2024

On the Essence and Prospect: An Investigation of Alignment Approaches for Big Models

Big models have achieved revolutionary breakthroughs in the field of AI, but they might also pose potential concerns. Addressing such concerns, alignment technologies were introduced to make these models conform to human preferences and values. Despite considerable advancements in the past year, various challenges lie in establishing the optimal alignment strategy, such as data cost and scalable oversight, and how to align remains an open question. In this survey paper, we comprehensively investigate value alignment approaches. We first unpack the historical context of alignment tracing back to the 1920s (where it comes from), then delve into the mathematical essence of alignment (what it is), shedding light on the inherent challenges. Following this foundation, we provide a detailed examination of existing alignment methods, which fall into three categories: Reinforcement Learning, Supervised Fine-Tuning, and In-context Learning, and demonstrate their intrinsic connections, strengths, and limitations, helping readers better understand this research area. In addition, two emerging topics, personal alignment, and multimodal alignment, are also discussed as novel frontiers in this field. Looking forward, we discuss potential alignment paradigms and how they could handle remaining challenges, prospecting where future alignment will go.

cs.AI↗

arXiv2024

Denevil: Towards Deciphering and Navigating the Ethical Values of Large Language Models via Instruction Learning

Large Language Models (LLMs) have made unprecedented breakthroughs, yet their increasing integration into everyday life might raise societal risks due to generated unethical content. Despite extensive study on specific issues like bias, the intrinsic values of LLMs remain largely unexplored from a moral philosophy perspective. This work delves into ethical values utilizing Moral Foundation Theory. Moving beyond conventional discriminative evaluations with poor reliability, we propose DeNEVIL, a novel prompt generation algorithm tailored to dynamically exploit LLMs' value vulnerabilities and elicit the violation of ethics in a generative manner, revealing their underlying value inclinations. On such a basis, we construct MoralPrompt, a high-quality dataset comprising 2,397 prompts covering 500+ value principles, and then benchmark the intrinsic values across a spectrum of LLMs. We discovered that most models are essentially misaligned, necessitating further ethical value alignment. In response, we develop VILMO, an in-context alignment method that substantially enhances the value compliance of LLM outputs by learning to generate appropriate value instructions, outperforming existing competitors. Our methods are suitable for black-box and open-source models, offering a promising initial step in studying the ethical values of LLMs.

cs.CL↗

arXiv2024

DiffSal: Joint Audio and Video Learning for Diffusion Saliency Prediction

Audio-visual saliency prediction can draw support from diverse modality complements, but further performance enhancement is still challenged by customized architectures as well as task-specific loss functions. In recent studies, denoising diffusion models have shown more promising in unifying task frameworks owing to their inherent ability of generalization. Following this motivation, a novel Diffusion architecture for generalized audio-visual Saliency prediction (DiffSal) is proposed in this work, which formulates the prediction problem as a conditional generative task of the saliency map by utilizing input audio and video as the conditions. Based on the spatio-temporal audio-visual features, an extra network Saliency-UNet is designed to perform multi-modal attention modulation for progressive refinement of the ground-truth saliency map from the noisy map. Extensive experiments demonstrate that the proposed DiffSal can achieve excellent performance across six challenging audio-visual benchmarks, with an average relative improvement of 6.3\% over the previous state-of-the-art results by six metrics.

cs.CV↗

arXiv2024

Two Ultracold Atoms in a Quasi-Two-Dimensional Box Confinement

We investigate the scattering and two-body bound states of two ultracold atoms in a quasi-two-dimensional (quasi-2D) confinement, with the confinement potential being an infinite square well (box potential) in the transverse ($z$-) direction, and the motion of the atoms in the $x$-$y$ plane being free. Specifically, we calculate the effective 2D scattering length and 2D effective range of the low-energy scattering, as well as the energy and the transverse-excited-mode probability of the bound states. Comparing these results with those obtained under a harmonic transverse confinement potential, we find that in most of the cases the 2D effective range for the box confinement is approximately 0.28 of the one for the harmonic confinement. Moreover, the transverse-excited-mode probability of the bound states for the box confinement is also much lower than the one for the harmonic confinement. These results suggest that the transverse excitation in the box confinement is notably weaker than the one in a harmonic confinement. Therefore, achieving quasi-2D ultracold gases well-described by pure-2D effective models, particularly those with 2D contact interaction, is more feasible through box confinement. Our results are helpful for the quantum simulation of 2D many-body physics with ultracold atoms, e.g., the suppression of 2D effective range may lead to an enhancement of quantum anomaly in two-dimensional Fermi Gases. Additionally, our calculation method is applicable to the two-body problems of ultracold atoms in other types of quasi-2D confinements.

cond-mat.quant-gas↗

arXiv2024

VN Network: Embedding Newly Emerging Entities with Virtual Neighbors

Embedding entities and relations into continuous vector spaces has attracted a surge of interest in recent years. Most embedding methods assume that all test entities are available during training, which makes it time-consuming to retrain embeddings for newly emerging entities. To address this issue, recent works apply the graph neural network on the existing neighbors of the unseen entities. In this paper, we propose a novel framework, namely Virtual Neighbor (VN) network, to address three key challenges. Firstly, to reduce the neighbor sparsity problem, we introduce the concept of the virtual neighbors inferred by rules. And we assign soft labels to these neighbors by solving a rule-constrained problem, rather than simply regarding them as unquestionably true. Secondly, many existing methods only use one-hop or two-hop neighbors for aggregation and ignore the distant information that may be helpful. Instead, we identify both logic and symmetric path rules to capture complex patterns. Finally, instead of one-time injection of rules, we employ an iterative learning scheme between the embedding method and virtual neighbor prediction to capture the interactions within. Experimental results on two knowledge graph completion tasks demonstrate that our VN network significantly outperforms state-of-the-art baselines. Furthermore, results on Subject/Object-R show that our proposed VN network is highly robust to the neighbor sparsity problem.

cs.LG↗

arXiv2024

HIP Network: Historical Information Passing Network for Extrapolation Reasoning on Temporal Knowledge Graph

In recent years, temporal knowledge graph (TKG) reasoning has received significant attention. Most existing methods assume that all timestamps and corresponding graphs are available during training, which makes it difficult to predict future events. To address this issue, recent works learn to infer future events based on historical information. However, these methods do not comprehensively consider the latent patterns behind temporal changes, to pass historical information selectively, update representations appropriately and predict events accurately. In this paper, we propose the Historical Information Passing (HIP) network to predict future events. HIP network passes information from temporal, structural and repetitive perspectives, which are used to model the temporal evolution of events, the interactions of events at the same time step, and the known events respectively. In particular, our method considers the updating of relation representations and adopts three scoring functions corresponding to the above dimensions. Experimental results on five benchmark datasets show the superiority of HIP network, and the significant improvements on Hits@1 prove that our method can more accurately predict what is going to happen.

cs.AI↗

arXiv2024

Online Physical Enhanced Residual Learning for Connected Autonomous Vehicles Platoon Centralized Control

This paper introduces an online physical enhanced residual learning (PERL) framework for Connected Autonomous Vehicles (CAVs) platoon, aimed at addressing the challenges posed by the dynamic and unpredictable nature of traffic environments. The proposed framework synergistically combines a physical model, represented by Model Predictive Control (MPC), with data-driven online Q-learning. The MPC controller, enhanced for centralized CAV platoons, employs vehicle velocity as a control input and focuses on multi-objective cooperative optimization. The learning-based residual controller enriches the MPC with prior knowledge and corrects residuals caused by traffic disturbances. The PERL framework not only retains the interpretability and transparency of physics-based models but also significantly improves computational efficiency and control accuracy in real-world scenarios. The experimental results present that the online Q-learning PERL controller, in comparison to the MPC controller and PERL controller with a neural network, exhibits significantly reduced position and velocity errors. Specifically, the PERL's cumulative absolute position and velocity errors are, on average, 86.73% and 55.28% lower than the MPC's, and 12.82% and 18.83% lower than the neural network-based PERL's, in four tests with different reference trajectories and errors. The results demonstrate our advanced framework's superior accuracy and quick convergence capabilities, proving its effectiveness in maintaining platoon stability under diverse conditions.

cs.MA↗

arXiv2024

Frequency-aware Graph Signal Processing for Collaborative Filtering

Graph Signal Processing (GSP) based recommendation algorithms have recently attracted lots of attention due to its high efficiency. However, these methods failed to consider the importance of various interactions that reflect unique user/item characteristics and failed to utilize user and item high-order neighborhood information to model user preference, thus leading to sub-optimal performance. To address the above issues, we propose a frequency-aware graph signal processing method (FaGSP) for collaborative filtering. Firstly, we design a Cascaded Filter Module, consisting of an ideal high-pass filter and an ideal low-pass filter that work in a successive manner, to capture both unique and common user/item characteristics to more accurately model user preference. Then, we devise a Parallel Filter Module, consisting of two low-pass filters that can easily capture the hierarchy of neighborhood, to fully utilize high-order neighborhood information of users/items for more accurate user preference modeling. Finally, we combine these two modules via a linear model to further improve recommendation accuracy. Extensive experiments on six public datasets demonstrate the superiority of our method from the perspectives of prediction accuracy and training efficiency compared with state-of-the-art GCN-based recommendation methods and GSP-based recommendation methods.

cs.IR↗

arXiv2024

Phonon and defect mediated quantum anomalous Hall insulator to metal transition in magnetically doped topological insulators

Quantum Anomalous Hall (QAH) state in six quintuple layer Cr$_{0.1}$(Bi$_{0.2}$Sb$_{0.8}$)$_{1.9}$Te$_3$ thin films were studied through scanning tunneling spectroscopy (STS) and electrical transport measurements. While the surface state is gapless above the Curie temperature ($T_\mathrm{C} \approx 30$ K), scanning tunneling spectroscopy (STS) of the sample reveals a topologically non-trivial gap with an average value of $\approx 13.5$ meV at 4.2 K below the ferromagnetic transition. Nonetheless, areal STS scans of the magnetic topological insulator exhibit energy modulations on the order of several meV's in the surface bands which result in the valence band maximum in some regions becoming higher than the energy of the conduction band minimum of some other regions that are spatially separated by no more than 3 nm. First principle calculations demonstrate that the origin of the observed inhomogeneous energy band alignment is an outcome of many-body interactions, namely electron-defect interactions and electron-phonon interactions. Defects play the role of locally modifying the energy landscape of surface bands while electron-phonon interactions renormalize the surface bands such that the surface gap becomes reduced by more than 1 meV as temperature is raised from 0 to 4.2 K. These many-body interactions at a finite temperature result in substantial increase of electron tunneling across the spatially separated conduction band pockets even for finite temperatures well below $T_\mathrm{C}$ , thus driving the magnetic topological insulator out of its QAH insulating phase into a metallic phase at a relatively low temperature.

cond-mat.str-el↗

arXiv2024

The Stellar "Snake" -- II: The Mass Function

We present a comprehensive investigation on the mass function (MF) of a snake-like stellar structure in the solar neighbourhood, building on our previous discovery. To ensure the reliability of the data, we reselect the member stars of the Stellar ``Snake'' in the latest {\it Gaia} Data Release 3 using the same approach as the initial series of articles. We also precisely measure the physical parameters of the clusters within the Stellar Snake. In light of the high completeness of the member stars in the cluster regions, we develop a simulated model color-magnitude diagram-based inference method to derive the mass function, binary fraction, and mass-ratio distribution of the clusters in the Stellar Snake. Notably, despite their similar ages and metallicity, we discover systematic variations in the MFs along the elongation direction of the Snake in the mass range of 0.5 to 2.0 M$_\odot$. The ``head'' of the Snake conforms to a canonical initial mass function with a power-law slope of $α\sim-2.3$. Extending towards the ``tail,'' the MF becomes more top-light, indicating a deficiency of massive stars within these clusters. This result provides evidence for the delayed formation of massive stars in the clusters. Such clues give support to the hypothesis that the Stellar Snake constitutes as a hierarchically primordial structure.

astro-ph.GA↗

arXiv2024

Observation of GRB 221009A early afterglow in X/$γ$-ray energy band

The early afterglow of a Gamma-ray burst (GRB) can provide critical information on the jet and progenitor of the GRB. The extreme brightness of GRB 221009A allows us to probe its early afterglow in unprecedented detail. In this letter, we report comprehensive observation results of the early afterglow of GRB 221009A (from $T_0$+660 s to $T_0$+1860 s, where $T_0$ is the \textit{Insight}-HXMT/HE trigger time) in X/$γ$-ray energy band (from 20 keV to 20 MeV) by \textit{Insight}-HXMT/HE, GECAM-C and \textit{Fermi}/GBM. We find that the spectrum of the early afterglow in 20 keV-20 MeV could be well described by a cutoff power-law with an extra power-law which dominates the low and high energy bands respectively. The cutoff power-law $E_{\rm peak}$ is $\sim$ 30 keV and the power-law photon index is $\sim$ 1.8 throughout the early afterglow phase. By fitting the light curves in different energy bands, we find that a significant achromatic break (from keV to TeV) is required at $T_0$ + 1246$^{+27}_{-26}$ s (i.e. 1021 s since the afterglow starting time $T_{\rm AG}$=$T_0$+225 s), providing compelling evidence of a jet break. Interestingly, both the pre-break and post-break decay slopes vary with energy, and these two slopes become closer in the lower energy band, making the break less identifiable. Intriguingly, the spectrum of the early afterglow experienced a slight hardening before the break and a softening after the break. These results provide new insights into the understanding of this remarkable GRB.

astro-ph.HE↗

arXiv2024

APAR: LLMs Can Do Auto-Parallel Auto-Regressive Decoding

The massive adoption of large language models (LLMs) demands efficient deployment strategies. However, the auto-regressive decoding process, which is fundamental to how most LLMs generate text, poses challenges to achieve efficient serving. In this work, we introduce a parallel auto-regressive generation method. By instruct-tuning on general domain data that contains hierarchical structures, we enable LLMs to independently plan their generation process and perform auto-parallel auto-regressive (APAR) generation, significantly reducing the number of generation steps. APAR alone can achieve up to 2x speed-up, and when combined with speculative decoding, the speed-up can reach up to 4x. In addition, APAR reduces the key-value cache consumption and attention computation during generation. This leads to a throughput increase of 20-70% and a latency reduce of 20-35% in high-throughput scenarios, compared to state-of-the-art serving frameworks.

cs.CL↗

arXiv2024

Early Warning Prediction with Automatic Labeling in Epilepsy Patients

Early warning for epilepsy patients is crucial for their safety and well-being, in particular to prevent or minimize the severity of seizures. Through the patients' EEG data, we propose a meta learning framework to improve the prediction of early ictal signals. The proposed bi-level optimization framework can help automatically label noisy data at the early ictal stage, as well as optimize the training accuracy of the backbone model. To validate our approach, we conduct a series of experiments to predict seizure onset in various long-term windows, with LSTM and ResNet implemented as the baseline models. Our study demonstrates that not only the ictal prediction accuracy obtained by meta learning is significantly improved, but also the resulting model captures some intrinsic patterns of the noisy data that a single backbone model could not learn. As a result, the predicted probability generated by the meta network serves as a highly effective early warning indicator.

cs.LG↗

arXiv2024

Re-evaluating the Memory-balanced Pipeline Parallelism: BPipe

Pipeline parallelism is an essential technique in the training of large-scale Transformer models. However, it suffers from imbalanced memory consumption, leading to insufficient memory utilization. The BPipe technique was proposed to address this issue and has proven effective in the GPT-3 model. Nevertheless, our experiments have not yielded similar benefits for LLaMA training. Additionally, BPipe only yields negligible benefits for GPT-3 training when applying flash attention. We analyze the underlying causes of the divergent performance of BPipe on GPT-3 and LLaMA. Furthermore, we introduce a novel method to estimate the performance of BPipe.

cs.LG↗