arXiv subjects

Hao Ding

Publications and source records attributed to Hao Ding.

At least 73 records · Page 4Linked to original sources

arXiv2024

Plug-and-Play Performance Estimation for LLM Services without Relying on Labeled Data

Large Language Model (LLM) services exhibit impressive capability on unlearned tasks leveraging only a few examples by in-context learning (ICL). However, the success of ICL varies depending on the task and context, leading to heterogeneous service quality. Directly estimating the performance of LLM services at each invocation can be laborious, especially requiring abundant labeled data or internal information within the LLM. This paper introduces a novel method to estimate the performance of LLM services across different tasks and contexts, which can be "plug-and-play" utilizing only a few unlabeled samples like ICL. Our findings suggest that the negative log-likelihood and perplexity derived from LLM service invocation can function as effective and significant features. Based on these features, we utilize four distinct meta-models to estimate the performance of LLM services. Our proposed method is compared against unlabeled estimation baselines across multiple LLM services and tasks. And it is experimentally applied to two scenarios, demonstrating its effectiveness in the selection and further optimization of LLM services.

cs.PF↗

arXiv2024

Pathfinding pulsar observations with the CVN incorporating the FAST

The importance of Very Long Baseline Interferometry (VLBI) for pulsar research is becoming increasingly prominent and receiving more and more attention. In this paper, we present pathfinding pulsar observation results with the Chinese VLBI Network (CVN) incorporating the Five-hundred-meter Aperture Spherical radio Telescope (FAST). On MJD 60045 (April 11th, 2023), PSRs B0919+06 and B1133+16 were observed with the phase-referencing mode in the L-band using four radio telescopes (FAST, TianMa, Haoping and Nanshan) and correlated with the pulsar binning mode of the distributed FX-style software correlator in Shanghai. After further data processing with the NRAO Astronomical Image Processing System (AIPS), we detected these two pulsars and fitted their current positions with accuracy at the milliarcsecond level. By comparison, our results show significantly better agreement with predicted values based on historical VLBI observations than that with previous timing observations, as pulsar astrometry with the VLBI provides a more direct and model-independent method for accurately obtaining related parameters.

astro-ph.HE↗

arXiv2024

A millisecond pulsar position determined to 0.2 milliarcsecond precision with VLBI

Precise millisecond pulsar (MSP) positions determined with very long baseline interferometry (VLBI) hold the key to building the connection between the kinematic and dynamic reference frames respectively used by VLBI and pulsar timing. The frame connection would provide an important pathway to examining the planetary ephemerides used in pulsar timing, and potentially enhancing the sensitivities of pulsar timing arrays used to detect stochastic gravitational-wave background at nano-Hz regime. We aim at significantly improving the VLBI-based MSP position from its current $\gtrsim1\,$mas precision level by reducing the two dominant components in the positional uncertainty -- the propagation-related uncertainty and the uncertainty resulting from the frequency-dependent core shifts of the reference sources. We introduce a new differential astrometry strategy of using multiple calibrators observed at several widely separated frequencies, which we call PINPT (Phase-screen Interpolation plus frequeNcy-dePendent core shifT correction; read as "pinpoint") for brevity. The strategy allows determination of the core-shift and mitigates the impact of residual delay in the atmosphere. We implemented the strategy on PSR J2222-0137, an MSP well constrained astrometrically with VLBI and pulsar timing. Using the PINPT strategy, we determined core shifts for 4 AGNs around PSR J2222-0137, and derived a VLBI-based pulsar position with uncertainty of 0.17 mas and 0.32 mas in right ascension and declination, respectively, approaching the uncertainty level of the best-determined timing-based MSP positions. The realization of the PINPT strategy promises a factor-of-5 positional precision enhancement (over conventional VLBI astrometry) for all kinds of compact radio sources observed at $\lesssim2$ GHz, including most fast radio bursts.

astro-ph.IM↗

arXiv2024

Performance and Non-adversarial Robustness of the Segment Anything Model 2 in Surgical Video Segmentation

Fully supervised deep learning (DL) models for surgical video segmentation have been shown to struggle with non-adversarial, real-world corruptions of image quality including smoke, bleeding, and low illumination. Foundation models for image segmentation, such as the segment anything model (SAM) that focuses on interactive prompt-based segmentation, move away from semantic classes and thus can be trained on larger and more diverse data, which offers outstanding zero-shot generalization with appropriate user prompts. Recently, building upon this success, SAM-2 has been proposed to further extend the zero-shot interactive segmentation capabilities from independent frame-by-frame to video segmentation. In this paper, we present a first experimental study evaluating SAM-2's performance on surgical video data. Leveraging the SegSTRONG-C MICCAI EndoVIS 2024 sub-challenge dataset, we assess SAM-2's effectiveness on uncorrupted endoscopic sequences and evaluate its non-adversarial robustness on videos with corrupted image quality simulating smoke, bleeding, and low brightness conditions under various prompt strategies. Our experiments demonstrate that SAM-2, in zero-shot manner, can achieve competitive or even superior performance compared to fully-supervised deep learning models on surgical video data, including under non-adversarial corruptions of image quality. Additionally, SAM-2 consistently outperforms the original SAM and its medical variants across all conditions. Finally, frame-sparse prompting can consistently outperform frame-wise prompting for SAM-2, suggesting that allowing SAM-2 to leverage its temporal modeling capabilities leads to more coherent and accurate segmentation compared to frequent prompting.

eess.IV↗

arXiv2024

Divide and Fuse: Body Part Mesh Recovery from Partially Visible Human Images

We introduce a novel bottom-up approach for human body mesh reconstruction, specifically designed to address the challenges posed by partial visibility and occlusion in input images. Traditional top-down methods, relying on whole-body parametric models like SMPL, falter when only a small part of the human is visible, as they require visibility of most of the human body for accurate mesh reconstruction. To overcome this limitation, our method employs a "Divide and Fuse (D&F)" strategy, reconstructing human body parts independently before fusing them, thereby ensuring robustness against occlusions. We design Human Part Parametric Models (HPPM) that independently reconstruct the mesh from a few shape and global-location parameters, without inter-part dependency. A specially designed fusion module then seamlessly integrates the reconstructed parts, even when only a few are visible. We harness a large volume of ground-truth SMPL data to train our parametric mesh models. To facilitate the training and evaluation of our method, we have established benchmark datasets featuring images of partially visible humans with HPPM annotations. Our experiments, conducted on these benchmark datasets, demonstrate the effectiveness of our D&F method, particularly in scenarios with substantial invisibility, where traditional approaches struggle to maintain reconstruction quality.

cs.CV↗

arXiv2024

VLBA Astrometry of the Fastest-spinning Magnetar Swift J1818.0-1607: A Large Trigonometric Distance & A Small Transverse Velocity

In addition to being the most magnetic objects in the known universe, magnetars are the only objects observed to generate fast-radio-burst-like emissions. The formation mechanism of magnetars is still highly debated, and may potentially be probed with the magnetar velocity distribution. We carried out a 3-year-long astrometric campaign on Swift J1818.0-1607 -- the fastest-spinning magnetar, using the Very Long Baseline Array. After applying the phase-calibrating 1D interpolation strategy, we obtained a small proper motion of 8.5 $\mathrm{mas~yr^{-1}}$ magnitude, and a parallax of $0.12\pm0.02$ mas (uncertainties at $1\,σ$ confidence throughout the Letter) for Swift J1818.0-1607. The latter is the second magnetar parallax, and is among the smallest neutron star parallaxes ever determined. From the parallax, we derived the distance $9.4^{+2.0}_{-1.6}$ kpc, which locates Swift J1818.0-1607 at the far side of the Galactic central region. Combined with the distance, the small proper motion leads to a transverse peculiar velocity $v_\perp=48^{+50}_{-16}$ $\mathrm{km~s^{-1}}$ -- a new lower limit to magnetar $v_\perp$. Incorporating previous $v_\perp$ estimates of seven other magnetars, we acquired $v_\perp=149^{+132}_{-68}$ $\mathrm{km~s^{-1}}$ for the sample of astrometrically studied magnetars, corresponding to the three-dimensional space velocity $\sim190^{+168}_{-87}$ $\mathrm{km~s^{-1}}$, smaller than the average level of young pulsars. Additionally, we found that the magnetar velocity sample does not follow the unimodal young pulsar velocity distribution reported by Hobbs et al. at $>2\,σ$ confidence, while loosely agreeing with more recent bimodal young pulsar velocity distributions derived from relatively small samples of quality astrometric determinations.

astro-ph.HE↗

arXiv2024

VLBA Astrometry of the Galactic Double Neutron Stars PSR J0509+3801 and PSR J1930-1852: A Preliminary Transverse Velocity Distribution of Double Neutron Stars and Its Implications

The mergers of double neutron stars (DNSs) systems are believed to drive the majority of short $γ$-ray bursts (SGRBs), while also serving as production sites of heavy r-process elements. Despite being key to i) confirming the nature of the extragalactic SGRBs, ii) addressing the poorly-understood r-process enrichment in the ultra-faint dwarf galaxies (UFDGs), and iii) probing the formation process of DNS systems, the space velocity distribution of DNSs is still poorly constrained due to the small number of DNSs with well-determined astrometry. In this work, we determine new proper motions and parallaxes of two Galactic DNSs -- PSR J0509+3801 and PSR J1930-1852, using the Very Long Baseline Array, and estimate the transverse velocities $v_\perp$ of all the 11 isolated Galactic DNSs having proper motion measurements in a consistent manner. Our correlation analysis reveals that the DNS $v_\perp$ is tentatively correlated with three parameters: spin period, orbital eccentricity, and companion mass. With the preliminary $v_\perp$ distribution, we obtain the following findings. Firstly, the refined $v_\perp$ distribution is confirmed to agree with the observed displacements of the localized SGRBs from their host galaxy birth sites. Secondly, we estimate that around 11% and 25% of DNSs remain gravitationally bound to UFDGs with escape velocities of 15$\mathrm{~km~s^{-1}}$ and 25$\mathrm{~km~s^{-1}}$, respectively. Hence, the retained DNSs might indeed be responsible for the r-process enrichment confirmed so far in a few UFDGs. Finally, we discuss how a future ensemble of astrometrically determined DNSs may probe the multimodality of the $v_\perp$ distribution.

astro-ph.HE↗

arXiv2024

VLBI Astrometry of Radio Stars to Link Radio and Optical Celestial Reference Frames: Observing Strategies

The Gaia celestial reference frame (Gaia-CRF) will benefit from a close assessment with independent methods, such as Very Long Baseline Interferometry (VLBI) measurements of radio stars at bright magnitudes. However, obtaining full astrometric parameters for each radio star through VLBI measurements demands a significant amount of observation time. This study proposes an efficient observing strategy that acquires double-epoch VLBI positions to measure the positions and proper motions of radio stars at a reduced cost. The solution for CRF link compatible with individual VLBI position measurements is introduced, and the optimized observing epoch scheduling is discussed. Applying this solution to observational data yields results sensitive to sample increase or decrease, yet they remain consistently in line with the literature at the 1-sigma level. This suggests the potential for improvement with a larger sample size. Simulations for adding observations demonstrate the double-epoch strategy reduces CRF link parameter uncertainties by over 30% compared to the five-parameter strategy.

astro-ph.IM↗

arXiv2024

LVDiffusor: Distilling Functional Rearrangement Priors from Large Models into Diffusor

Object rearrangement, a fundamental challenge in robotics, demands versatile strategies to handle diverse objects, configurations, and functional needs. To achieve this, the AI robot needs to learn functional rearrangement priors in order to specify precise goals that meet the functional requirements. Previous methods typically learn such priors from either laborious human annotations or manually designed heuristics, which limits scalability and generalization. In this work, we propose a novel approach that leverages large models to distill functional rearrangement priors. Specifically, our approach collects diverse arrangement examples using both LLMs and VLMs and then distills the examples into a diffusion model. During test time, the learned diffusion model is conditioned on the initial configuration and guides the positioning of objects to meet functional requirements. In this manner, we create a handshaking point that combines the strengths of conditional generative models and large models. Extensive experiments on multiple domains, including real-world scenarios, demonstrate the effectiveness of our approach in generating compatible goals for object rearrangement tasks, significantly outperforming baseline methods.

cs.RO↗

arXiv2024

DaReNeRF: Direction-aware Representation for Dynamic Scenes

Addressing the intricate challenge of modeling and re-rendering dynamic scenes, most recent approaches have sought to simplify these complexities using plane-based explicit representations, overcoming the slow training time issues associated with methods like Neural Radiance Fields (NeRF) and implicit representations. However, the straightforward decomposition of 4D dynamic scenes into multiple 2D plane-based representations proves insufficient for re-rendering high-fidelity scenes with complex motions. In response, we present a novel direction-aware representation (DaRe) approach that captures scene dynamics from six different directions. This learned representation undergoes an inverse dual-tree complex wavelet transformation (DTCWT) to recover plane-based information. DaReNeRF computes features for each space-time point by fusing vectors from these recovered planes. Combining DaReNeRF with a tiny MLP for color regression and leveraging volume rendering in training yield state-of-the-art performance in novel view synthesis for complex dynamic scenes. Notably, to address redundancy introduced by the six real and six imaginary direction-aware wavelet coefficients, we introduce a trainable masking approach, mitigating storage issues without significant performance decline. Moreover, DaReNeRF maintains a 2x reduction in training time compared to prior art while delivering superior performance.

cs.CV↗

arXiv2024

Logic-Scaffolding: Personalized Aspect-Instructed Recommendation Explanation Generation using LLMs

The unique capabilities of Large Language Models (LLMs), such as the natural language text generation ability, position them as strong candidates for providing explanation for recommendations. However, despite the size of the LLM, most existing models struggle to produce zero-shot explanations reliably. To address this issue, we propose a framework called Logic-Scaffolding, that combines the ideas of aspect-based explanation and chain-of-thought prompting to generate explanations through intermediate reasoning steps. In this paper, we share our experience in building the framework and present an interactive demonstration for exploring our results.

cs.AI↗

arXiv2024

Pre-trained Recommender Systems: A Causal Debiasing Perspective

Recent studies on pre-trained vision/language models have demonstrated the practical benefit of a new, promising solution-building paradigm in AI where models can be pre-trained on broad data describing a generic task space and then adapted successfully to solve a wide range of downstream tasks, even when training data is severely limited (e.g., in zero- or few-shot learning scenarios). Inspired by such progress, we investigate in this paper the possibilities and challenges of adapting such a paradigm to the context of recommender systems, which is less investigated from the perspective of pre-trained model. In particular, we propose to develop a generic recommender that captures universal interaction patterns by training on generic user-item interaction data extracted from different domains, which can then be fast adapted to improve few-shot learning performance in unseen new domains (with limited data). However, unlike vision/language data which share strong conformity in the semantic space, universal patterns underlying recommendation data collected across different domains (e.g., different countries or different E-commerce platforms) are often occluded by both in-domain and cross-domain biases implicitly imposed by the cultural differences in their user and item bases, as well as their uses of different e-commerce platforms. As shown in our experiments, such heterogeneous biases in the data tend to hinder the effectiveness of the pre-trained model. To address this challenge, we further introduce and formalize a causal debiasing perspective, which is substantiated via a hierarchical Bayesian deep learning model, named PreRec. Our empirical studies on real-world data show that the proposed model could significantly improve the recommendation performance in zero- and few-shot learning settings under both cross-market and cross-platform scenarios.

cs.IR↗

arXiv2024

Consistent Intradecadal/Interdecadal Oscillations in the Surface Geomagnetic Observations and in the $Δ$LOD: New Findings and Unresolved Problems

Surface geomagnetic observations and length of day variations (dLOD) have played an important role in interpreting the long-period motions of the Earth's core. Focusing on the 5-30yr period band, we use the optimal sequence estimation method to analyze the global geomagnetic vertical observations, and newly find that the about 6yr/8.6yr signals, the about7.6yr/13.6yr/22.5yr signals, the about 15.4yr and about 18.6yr signals respectively have the Y2,2, Y2,-2, Y2,-1 and Y2,0 spatial patterns; in which, the about 7.6yr/8.6yr/13.6yr/15.4yr/18.6yr signals are clearly detected for the first time. We also find that the five Y2,+/-2-related "equivalent" excitation signals have good phase consistency with the corresponding signals in the dLOD, and they have the same about 0.042 amplitude scaling factor. Determining the physical mechanisms behind those seven signals still leaves many unresolved problems, and our new findings make those problems more complicated. As a preliminary thought, we suggest the five Y2,+/-2-related signals may be originated from the Magnetic-Archimedes-Coriolis waves

physics.geo-ph↗

arXiv2023

Inner core static tilt inferred from intradecadal oscillation in the Earth's rotation

The geodynamic state of the inner core remains an enigma, encompassing the presence of a static tilt between the inner core and mantle. Following the experimental confirmation of an ~8.5yr signal in polar motion as the inner core wobble (ICW), a normal mode of the inner core, we report that the ~8.5yr oscillation contained in the length-of-day variations in the Earth's rotation has good phase consistency with it. Our analysis demonstrated a 0.17° static tilt of the inner core (more likely towards ~90°W) relative to the mantle, which is two orders of magnitude lower than the 10° assumed in certain geodynamic researches. This tilt is consistent with the assumption that the average density in the northwestern hemisphere of the inner core should be greater than that in the other regions. Besides, the observed ICW period (8.5yr) suggests a 0.52g/cm3 density jump at the inner core boundary.

physics.geo-ph↗

arXiv2023

Stochastic differential equations and stochastic parallel translations in the Wasserstein space

We will develop some elements in stochastic analysis in the Wasserstein space $\mathbb{P}_2(M)$ over a compact Riemannian manifold $M$, such as intrinsic It$ô$ formulae, stochastic regular curves and parallel translations along them. We will establish the existence of parallel translations along regular curves, or stochastic regular curves in case of $\mathbb{P}_2(\mathbb{T})$. Surprisingly enough, in this last case, the equation defining stochastic parallel translations is a SDE on a Hilbert space, instead of a SPDE.

math.PR↗

arXiv2023

Personalized Federated Domain Adaptation for Item-to-Item Recommendation

Item-to-Item (I2I) recommendation is an important function in most recommendation systems, which generates replacement or complement suggestions for a particular item based on its semantic similarities to other cataloged items. Given that subsets of items in a recommendation system might be co-interacted with by the same set of customers, graph-based models, such as graph neural networks (GNNs), provide a natural framework to combine, ingest and extract valuable insights from such high-order relational interactions between cataloged items, as well as their metadata features, as has been shown in many recent studies. However, learning GNNs effectively for I2I requires ingesting a large amount of relational data, which might not always be available, especially in new, emerging market segments. To mitigate this data bottleneck, we postulate that recommendation patterns learned from existing mature market segments (with private data) could be adapted to build effective warm-start models for emerging ones. To achieve this, we propose and investigate a personalized federated modeling framework based on GNNs to summarize, assemble and adapt recommendation patterns across market segments with heterogeneous customer behaviors into effective local models. Our key contribution is a personalized graph adaptation model that bridges the gap between recent literature on federated GNNs and (non-graph) personalized federated learning, which either does not optimize for the adaptability of the federated model or is restricted to local models with homogeneous parameterization, excluding GNNs with heterogeneous local graphs.

cs.IR↗

arXiv2023

GCNSLIM: Graph Convolutional Network with Sparse Linear Methods for E-government Service Recommendation

Graph Convolutional Networks have made significant strides in Collabora-tive Filtering recommendations. However, existing GCN-based CF methods are mainly based on matrix factorization and incorporate some optimization tech-niques to enhance performance, which are not enough to handle the complexities of diverse real-world recommendation scenarios. E-government service recommendation is a crucial area for recommendation re-search as it involves rigid aspects of people's lives. However, it has not received ad-equate attention in comparison to other recommendation scenarios like news and music recommendation. We empirically find that when existing GCN-based CF methods are directly applied to e-government service recommendation, they are limited by the MF framework and showing poor performance. This is because MF's equal treatment of users and items is not appropriate for scenarios where the number of users and items is unbalanced. In this work, we propose a new model, GCNSLIM, which combines GCN and sparse linear methods instead of combining GCN and MF to accommodate e-government service recommendation. In particular, GCNSLIM explicitly injects high-order collaborative signals obtained from multi-layer light graph convolutions into the item similarity matrix in the SLIM frame-work, effectively improving the recommendation accuracy. In addition, we propose two optimization measures, removing layer 0 embedding and adding nonlinear acti-vation, to further adapt to the characteristics of e-government service recommenda-tion scenarios. Furthermore, we propose a joint optimization mode to adapt to more diverse recommendation scenarios. We conduct extensive experiments on a real e-government service dataset and a common public dataset and demonstrate the ef-fectiveness of GCNSLIM in recommendation accuracy and operational performance.

cs.IR↗

arXiv2023

Twin-S: A Digital Twin for Skull-base Surgery

Purpose: Digital twins are virtual interactive models of the real world, exhibiting identical behavior and properties. In surgical applications, computational analysis from digital twins can be used, for example, to enhance situational awareness. Methods: We present a digital twin framework for skull-base surgeries, named Twin-S, which can be integrated within various image-guided interventions seamlessly. Twin-S combines high-precision optical tracking and real-time simulation. We rely on rigorous calibration routines to ensure that the digital twin representation precisely mimics all real-world processes. Twin-S models and tracks the critical components of skull-base surgery, including the surgical tool, patient anatomy, and surgical camera. Significantly, Twin-S updates and reflects real-world drilling of the anatomical model in frame rate. Results: We extensively evaluate the accuracy of Twin-S, which achieves an average 1.39 mm error during the drilling process. We further illustrate how segmentation masks derived from the continuously updated digital twin can augment the surgical microscope view in a mixed reality setting, where bone requiring ablation is highlighted to provide surgeons additional situational awareness. Conclusion: We present Twin-S, a digital twin environment for skull-base surgery. Twin-S tracks and updates the virtual model in real-time given measurements from modern tracking technologies. Future research on complementing optical tracking with higher-precision vision-based approaches may further increase the accuracy of Twin-S.

cs.HC↗