Searcharxiv⌕ Search

arXiv subjects

Bo Zhang

Publications and source records attributed to Bo Zhang.

At least 415 records · Page 23Linked to original sources
arXiv2024

DreamCom: Finetuning Text-guided Inpainting Model for Image Composition

The goal of image composition is merging a foreground object into a background image to obtain a realistic composite image. Recently, generative composition methods are built on large pretrained diffusion models, due to their unprecedented image generation ability. However, they are weak in preserving the foreground object details. Inspired by recent text-to-image generation customized for certain object, we propose DreamCom by treating image composition as text-guided image inpainting customized for certain object. Specifically , we finetune pretrained text-guided image inpainting model based on a few reference images containing the same object, during which the text prompt contains a special token associated with this object. Then, given a new background, we can insert this object into the background with the text prompt containing the special token. In practice, the inserted object may be adversely affected by the background, so we propose masked attention mechanisms to avoid negative background interference. Experimental results on DreamEditBench and our contributed MureCom dataset show the outstanding performance of our DreamCom.

cs.CV↗
arXiv2024

MLAD: A Unified Model for Multi-system Log Anomaly Detection

In spite of the rapid advancements in unsupervised log anomaly detection techniques, the current mainstream models still necessitate specific training for individual system datasets, resulting in costly procedures and limited scalability due to dataset size, thereby leading to performance bottlenecks. Furthermore, numerous models lack cognitive reasoning capabilities, posing challenges in direct transferability to similar systems for effective anomaly detection. Additionally, akin to reconstruction networks, these models often encounter the "identical shortcut" predicament, wherein the majority of system logs are classified as normal, erroneously predicting normal classes when confronted with rare anomaly logs due to reconstruction errors. To address the aforementioned issues, we propose MLAD, a novel anomaly detection model that incorporates semantic relational reasoning across multiple systems. Specifically, we employ Sentence-bert to capture the similarities between log sequences and convert them into highly-dimensional learnable semantic vectors. Subsequently, we revamp the formulas of the Attention layer to discern the significance of each keyword in the sequence and model the overall distribution of the multi-system dataset through appropriate vector space diffusion. Lastly, we employ a Gaussian mixture model to highlight the uncertainty of rare words pertaining to the "identical shortcut" problem, optimizing the vector space of the samples using the maximum expectation model. Experiments on three real-world datasets demonstrate the superiority of MLAD.

cs.SE↗
arXiv2024

The ALMA-QUARKS survey: Detection of two extremely dense substructures in a massive prestellar core

Only a handful of massive starless core candidates have been discovered so far, but none of them have been fully confirmed. Within the MM1 clump in the filamentary infrared dark cloud G34.43+0.24 that was covered by the ALMA-ATOMS survey at Band 3 ($\sim2\arcsec$, 6000\,au) and the ALMA-QUARKS survey at Band 6 ($\sim 0.3\arcsec$, 900\,au), two prestellar core candidates MM1-C and E1 with masses of 71 and 20 \solarmass~and radii of 2100--4400\,au were discovered. The two cores show no obvious sign of star-formation activities. In particular, MM1-C is a very promising massive prestellar core candidate with a total gas mass of 71\,\solarmass. Within MM1-C, we detected two extremely dense substructures, C1 and C2, as characterized by their high densities of $\rm n_{H_2}\sim 10^{8-9} cm^{-3}$. Moreover, evidence of further fragmentation in C2 was also revealed. We have detected the primordial fragmentation in the earliest stage of massive star formation, and we speculate that MM1-C would be the birthplace of a massive multiple system. However, we cannot fully rule out the possibility that the massive prestellar core MM1-C will just form a cluster of low-mass stars if it undergoes further fragmentation.

astro-ph.GA↗
arXiv2024

LogFormer: A Pre-train and Tuning Pipeline for Log Anomaly Detection

Log anomaly detection is a key component in the field of artificial intelligence for IT operations (AIOps). Considering log data of variant domains, retraining the whole network for unknown domains is inefficient in real industrial scenarios. However, previous deep models merely focused on extracting the semantics of log sequences in the same domain, leading to poor generalization on multi-domain logs. To alleviate this issue, we propose a unified Transformer-based framework for Log anomaly detection (LogFormer) to improve the generalization ability across different domains, where we establish a two-stage process including the pre-training and adapter-based tuning stage. Specifically, our model is first pre-trained on the source domain to obtain shared semantic knowledge of log data. Then, we transfer such knowledge to the target domain via shared parameters. Besides, the Log-Attention module is proposed to supplement the information ignored by the log-paring. The proposed method is evaluated on three public and one real-world datasets. Experimental results on multiple benchmarks demonstrate the effectiveness of our LogFormer with fewer trainable parameters and lower training costs.

cs.LG↗
arXiv2024

An Event-Oriented Diffusion-Refinement Method for Sparse Events Completion

Event cameras or dynamic vision sensors (DVS) record asynchronous response to brightness changes instead of conventional intensity frames, and feature ultra-high sensitivity at low bandwidth. The new mechanism demonstrates great advantages in challenging scenarios with fast motion and large dynamic range. However, the recorded events might be highly sparse due to either limited hardware bandwidth or extreme photon starvation in harsh environments. To unlock the full potential of event cameras, we propose an inventive event sequence completion approach conforming to the unique characteristics of event data in both the processing stage and the output form. Specifically, we treat event streams as 3D event clouds in the spatiotemporal domain, develop a diffusion-based generative model to generate dense clouds in a coarse-to-fine manner, and recover exact timestamps to maintain the temporal resolution of raw data successfully. To validate the effectiveness of our method comprehensively, we perform extensive experiments on three widely used public datasets with different spatial resolutions, and additionally collect a novel event dataset covering diverse scenarios with highly dynamic motions and under harsh illumination. Besides generating high-quality dense events, our method can benefit downstream applications such as object classification and intensity frame reconstruction.

cs.CV↗
arXiv2024

Outer-space branch-and-bound algorithm for generalized linear multiplicative programs

This paper introduces a new global optimization algorithm for solving the generalized linear multiplicative problem (GLMP). The algorithm starts by introducing $\bar{p}$ new variables and applying a logarithmic transformation to convert the problem into an equivalent problem (EP). By using the strong duality of linear program, a new convex relaxation subproblem is formulated to obtain the lower bounds for the optimal value of EP. This relaxation subproblem, combined with a simplicial branching process, forms the foundation of a simplicial branch-and-bound algorithm that can globally solve the problem. The paper also includes an analysis of the theoretical convergence and computational complexity of the algorithm. Additionally, numerical experiments are conducted to demonstrate the effectiveness of the proposed algorithm in various test instances.

math.OC↗
arXiv2023

MobileVLM : A Fast, Strong and Open Vision Language Assistant for Mobile Devices

We present MobileVLM, a competent multimodal vision language model (MMVLM) targeted to run on mobile devices. It is an amalgamation of a myriad of architectural designs and techniques that are mobile-oriented, which comprises a set of language models at the scale of 1.4B and 2.7B parameters, trained from scratch, a multimodal vision model that is pre-trained in the CLIP fashion, cross-modality interaction via an efficient projector. We evaluate MobileVLM on several typical VLM benchmarks. Our models demonstrate on par performance compared with a few much larger models. More importantly, we measure the inference speed on both a Qualcomm Snapdragon 888 CPU and an NVIDIA Jeston Orin GPU, and we obtain state-of-the-art performance of 21.5 tokens and 65.3 tokens per second, respectively. Our code will be made available at: https://github.com/Meituan-AutoML/MobileVLM.

cs.CV↗
arXiv2023

Generalizing the intention-to-treat effect of an active control against placebo from historical placebo-controlled trials to an active-controlled trial: A case study of the efficacy of daily oral TDF/FTC in the HPTN 084 study

In many clinical settings, an active-controlled trial design (e.g., a non-inferiority or superiority design) is often used to compare an experimental medicine to an active control (e.g., an FDA-approved, standard therapy). One prominent example is a recent phase 3 efficacy trial, HIV Prevention Trials Network Study 084 (HPTN 084), comparing long-acting cabotegravir, a new HIV pre-exposure prophylaxis (PrEP) agent, to the FDA-approved daily oral tenofovir disoproxil fumarate plus emtricitabine (TDF/FTC) in a population of heterosexual women in 7 African countries. One key complication of interpreting study results in an active-controlled trial like HPTN 084 is that the placebo arm is not present and the efficacy of the active control (and hence the experimental drug) compared to the placebo can only be inferred by leveraging other data sources. \bz{In this article, we study statistical inference for the intention-to-treat (ITT) effect of the active control using relevant historical placebo-controlled trials data under the potential outcomes (PO) framework}. We highlight the role of adherence and unmeasured confounding, discuss in detail identification assumptions and two modes of inference (point versus partial identification), propose estimators under identification assumptions permitting point identification, and lay out sensitivity analyses needed to relax identification assumptions. We applied our framework to estimating the intention-to-treat effect of daily oral TDF/FTC versus placebo in HPTN 084 using data from an earlier Phase 3, placebo-controlled trial of daily oral TDF/FTC (Partners PrEP).

stat.ME↗
arXiv2023

Towards Knowledge-driven Autonomous Driving

This paper explores the emerging knowledge-driven autonomous driving technologies. Our investigation highlights the limitations of current autonomous driving systems, in particular their sensitivity to data bias, difficulty in handling long-tail scenarios, and lack of interpretability. Conversely, knowledge-driven methods with the abilities of cognition, generalization and life-long learning emerge as a promising way to overcome these challenges. This paper delves into the essence of knowledge-driven autonomous driving and examines its core components: dataset \& benchmark, environment, and driver agent. By leveraging large language models, world models, neural rendering, and other advanced artificial intelligence techniques, these components collectively contribute to a more holistic, adaptive, and intelligent autonomous driving system. The paper systematically organizes and reviews previous research efforts in this area, and provides insights and guidance for future research and practical applications of autonomous driving. We will continually share the latest updates on cutting-edge developments in knowledge-driven autonomous driving along with the relevant valuable open-source resources at: \url{https://github.com/PJLab-ADG/awesome-knowledge-driven-AD}.

cs.RO↗
arXiv2023

Direct observation of topological magnon polarons in a multiferroic material

Magnon polarons are novel elementary excitations possessing hybrid magnonic and phononic signatures, and are responsible for many exotic spintronic and magnonic phenomena. Despite long-term sustained experimental efforts in chasing for magnon polarons, direct spectroscopic evidence of their existence is hardly observed. Here, we report the direct observation of magnon polarons using neutron spectroscopy on a multiferroic Fe$_{2}$Mo$_{3}$O$_{8}$ possessing strong magnon-phonon coupling. Specifically, below the magnetic ordering temperature, a gap opens at the nominal intersection of the original magnon and phonon bands, leading to two separated magnon-polaron bands. Each of the bands undergoes mixing, interconverting and reversing between its magnonic and phononic components. We attribute the formation of magnon polarons to the strong magnon-phonon coupling induced by Dzyaloshinskii-Moriya interaction. Intriguingly, we find that the band-inverted magnon polarons are topologically nontrivial. These results uncover exotic elementary excitations arising from the magnon-phonon coupling, and offer a new route to topological states by considering hybridizations between different types of fundamental excitations.

cond-mat.str-el↗
arXiv2023

Observation of a 1/3 Magnetisation Plateau Phase as Evidence for the Kitaev Interaction in a Honeycomb-Lattice Antiferromagnet

Fractional magnetisation plateaus, in which the magnetisation is pinned at a fraction of its saturated value within a range of external magnetic field, are spectacular macroscopic manifestations of the collective quantum behaviours. One prominent example of the plateau phase is found in spin-1/2 triangular-lattice antiferromagnets featuring strong geometrical frustration, and is often interpreted as quantum-fluctuation-stabilised state in magnetic field via the "order-by-disorder" mechanism. Here, we observe an unprecedented 1/3 magnetisation plateau between 5.2 and 7.4 T at 2 K in a spin-1 antiferromagnet Na$_3$Ni$_2$BiO$_6$ with a honeycomb lattice, where conventionally no geometrical frustration is anticipated. By carrying out elastic neutron scattering measurements, we propose the spin structure of the plateau phase to be an unusual partial spin-flop ferrimagnetic order, transitioning from the zigzag antiferromagnetic order in zero field. Our theoretical calculations show that the plateau phase is stabilised by the bond-anisotropic Kitaev interaction. These results provide a new paradigm for the exploration of rich quantum phases in frustrated magnets and exotic Kitaev physics in high-spin systems.

cond-mat.str-el↗
arXiv2023

Uniform far-field asymptotics of the two-layered Green function in 2D and application to wave scattering in a two-layered medium

In this paper, we establish new results for the uniform far-field asymptotics of the two-layered Green function (together with its derivatives) in 2D in the frequency domain. To the best of our knowledge, our results are the sharpest yet obtained. The steepest descent method plays an important role in the proofs of our results. Further, as an application of our new results, we derive the uniform far-field asymptotics of the scattered field to the acoustic scattering problem by buried obstacles in a two-layered medium with a locally rough interface. The results obtained in this paper provide a theoretical foundation for our recent work, where direct imaging methods have been developed to image the locally rough interface from phaseless total-field data or phased far-field data at a fixed frequency. It is believed that the results obtained in this paper will also be useful on its own right.

math.AP↗
arXiv2023

Tailoring coherent microwave emission from a solid-state hybrid system for room-temperature microwave quantum electronics

Quantum electronics operating in the microwave domain are burgeoning and becoming essential building blocks of quantum computers, sensors and communication devices. However, the field of microwave quantum electronics has long been dominated by the need for cryogenic conditions to maintain the delicate quantum characteristics. Here we report on a solid-state hybrid system, constituted by a photo-excited pentacene triplet spin ensemble coupled to a dielectric resonator, that is for the first time capable of both coherent microwave quantum amplification and oscillation at X band via the masing process at room temperature. By incorporating external driving and active dissipation control into the hybrid system, we achieve efficient tuning of the maser emission characteristics at around 9.4 GHz, which is key to optimizing the performance of the maser device. Our work not only pushes the boundaries of the operating frequency and functionality of the existing pentacene masers, but also demonstrate a universal route for controlling the masing process at room temperature, highlighting opportunities for optimizing emerging solid-state masers for quantum information processing and communication.

quant-ph↗
arXiv2023

A New Global Optimization Method Based on Simplex Branching for Solving a Class of Non-Convex QCQP Problems

Quadratic constrained quadratic programming problems often occur in various fields such as engineering practice, management science, and network communication. This article mainly studies a non convex quadratic programming problem with convex quadratic constraints. Firstly, based on our existing results, the problem is reconstructed as an equivalent problem with a simple concave quadratic objective function in the result space, with a convex feasible domain. A global optimization algorithm for solving equivalent problems is proposed based on a branch and bound framework that can ensure the global optimality of the solution. This algorithm combines effective relaxation processes with branching processes related to new external approximation techniques. Finally, the theoretical feasibility of the algorithm was analyzed.

math.OC↗
arXiv2023

Vortex $γ$ photon generation via spin-to-orbital angular momentum transfer in nonlinear Compton scattering

Vortex $γ$ photons with intrinsic orbital angular momenta (OAM) possess a wealth of applications in various fields, e.g.-strong-laser physics, nuclear physics, particle physics and astrophysics-yet their generation remains unsettled. In this work, we investigate the generation of vortex $γ$ photons via nonlinear Compton scattering of ultrarelativistic electrons in a circularly polarized laser pulse. We develop a quantum electrodynamics scattering theory that explicitly addresses the multiphoton absorption and the angular momentum transfer mechanism. In pulsed laser fields, we unveil the vortex phase structure of the scattering matrix element, discuss how the vortex phase could be transferred to the radiated photon, and derive the radiation rate of the vortex $γ$ photon. We numerically examine the energy spectra and beam characteristics of the radiation, while also investigating the influence of finite laser pulses on the angular momentum and energy distribution of the emitted vortex $γ$ photons.

hep-ph↗
arXiv2023

Calibration of metallicity of LAMOST M dwarf stars Using FGK+M wide binaries

Estimating precise metallicity of M dwarfs is a well-known difficult problem due to their complex spectra. In this work, we empirically calibrate the metallicity using wide binaries with a F, G, or K dwarf and a M dwarf companion. With 1308 FGK+M wide binaries well observed by LAMOST, we calibrated M dwarf's [Fe/H] by using the Stellar LAbel Machine (SLAM) model, a data-driven method based on support vector regression (SVR). The [Fe/H] labels of the training data are from FGK companions in range of [-1,0.5] dex. The Teffs are selected from Li et al. (2021), spanning [3100,4400] K. The uncertainties in SLAM estimates of [Fe/H] and Teff are ~0.15 dex and ~40 K, respectively, at snri > 100, where snri is the signal-to-noise ratio (SNR) at i-band of M dwarf spectra. We applied the trained SLAM model to determine the [Fe/H] and Teff for ~630,000 M dwarfs with low-resolution spectra in LAMOST DR9. Compared to other literature also using FGK+M wide binaries for calibration, our [Fe/H] estimates show no bias but a scatter of ~ 0.14-0.18 dex. However, the [Fe/H] compared to APOGEE shows a systematic difference of ~ 0.10-0.15 dex with a scatter of ~ 0.15-0.20 dex. While the Teff compared to APOGEE has a bias of 3 K with a scatter of 62 K, it is systematically higher by 180 K compared to other calibrations based on the bolometric temperature. Finally, we calculated the zeta index for 1308 M dwarf secondaries and presents a moderate correlation between zeta and [Fe/H].

astro-ph.SR↗
arXiv2023

FastPillars: A Deployment-friendly Pillar-based 3D Detector

The deployment of 3D detectors strikes one of the major challenges in real-world self-driving scenarios. Existing BEV-based (i.e., Bird Eye View) detectors favor sparse convolutions (known as SPConv) to speed up training and inference, which puts a hard barrier for deployment, especially for on-device applications. In this paper, to tackle the challenge of efficient 3D object detection from an industry perspective, we devise a deployment-friendly pillar-based 3D detector, termed FastPillars. First, we introduce a novel lightweight Max-and-Attention Pillar Encoding (MAPE) module specially for enhancing small 3D objects. Second, we propose a simple yet effective principle for designing a backbone in pillar-based 3D detection. We construct FastPillars based on these designs, achieving high performance and low latency without SPConv. Extensive experiments on two large-scale datasets demonstrate the effectiveness and efficiency of FastPillars for on-device 3D detection regarding both performance and speed. Specifically, FastPillars delivers state-of-the-art accuracy on Waymo Open Dataset with 1.8X speed up and 3.8 mAPH/L2 improvement over CenterPoint (SPConv-based). Our code is publicly available at: https://github.com/StiphyJay/FastPillars.

cs.CV↗
arXiv2023

Norm Tweaking: High-performance Low-bit Quantization of Large Language Models

As the size of large language models (LLMs) continues to grow, model compression without sacrificing accuracy has become a crucial challenge for deployment. While some quantization methods, such as GPTQ, have made progress in achieving acceptable 4-bit weight-only quantization, attempts at lower-bit quantization often result in severe performance degradation. In this paper, we introduce a technique called norm tweaking, which can be used as a plugin in current PTQ methods to achieve high precision while being cost-efficient. Our approach is inspired by the observation that rectifying the quantized activation distribution to match its float counterpart can readily restore accuracy for LLMs. To achieve this, we carefully design a tweaking strategy that includes calibration data generation and channel-wise distance constraint to update the weights of normalization layers for better generalization. We conduct extensive experiments on various datasets using several open-sourced LLMs. Our method demonstrates significant improvements in both weight-only quantization and joint quantization of weights and activations, surpassing existing PTQ methods. On GLM-130B and OPT-66B, our method even achieves the same level of accuracy at 2-bit quantization as their float ones. Our simple and effective approach makes it more practical for real-world applications.

cs.LG↗