arXiv subjects

Yi Yang

Publications and source records attributed to Yi Yang.

At least 559 records · Page 31Linked to original sources

arXiv2024

Understanding Privacy Risks of Embeddings Induced by Large Language Models

Large language models (LLMs) show early signs of artificial general intelligence but struggle with hallucinations. One promising solution to mitigate these hallucinations is to store external knowledge as embeddings, aiding LLMs in retrieval-augmented generation. However, such a solution risks compromising privacy, as recent studies experimentally showed that the original text can be partially reconstructed from text embeddings by pre-trained language models. The significant advantage of LLMs over traditional pre-trained models may exacerbate these concerns. To this end, we investigate the effectiveness of reconstructing original knowledge and predicting entity attributes from these embeddings when LLMs are employed. Empirical findings indicate that LLMs significantly improve the accuracy of two evaluated tasks over those from pre-trained models, regardless of whether the texts are in-distribution or out-of-distribution. This underscores a heightened potential for LLMs to jeopardize user privacy, highlighting the negative consequences of their widespread use. We further discuss preliminary strategies to mitigate this risk.

cs.CL↗

arXiv2024

Deep SE(3)-Equivariant Geometric Reasoning for Precise Placement Tasks

Many robot manipulation tasks can be framed as geometric reasoning tasks, where an agent must be able to precisely manipulate an object into a position that satisfies the task from a set of initial conditions. Often, task success is defined based on the relationship between two objects - for instance, hanging a mug on a rack. In such cases, the solution should be equivariant to the initial position of the objects as well as the agent, and invariant to the pose of the camera. This poses a challenge for learning systems which attempt to solve this task by learning directly from high-dimensional demonstrations: the agent must learn to be both equivariant as well as precise, which can be challenging without any inductive biases about the problem. In this work, we propose a method for precise relative pose prediction which is provably SE(3)-equivariant, can be learned from only a few demonstrations, and can generalize across variations in a class of objects. We accomplish this by factoring the problem into learning an SE(3) invariant task-specific representation of the scene and then interpreting this representation with novel geometric reasoning layers which are provably SE(3) equivariant. We demonstrate that our method can yield substantially more precise placement predictions in simulated placement tasks than previous methods trained with the same amount of data, and can accurately represent relative placement relationships data collected from real-world demonstrations. Supplementary information and videos can be found at https://sites.google.com/view/reldist-iclr-2023.

cs.RO↗

arXiv2024

Multiple charge-density-wave gaps in LaSbTe and CeSbTe as revealed by ultrafast spectroscopy

Utilizing ultrafast time-resolved pump-probe spectroscopy measurements, we investigate the photoinduced quasiparticle dynamics of the topological materials LaSbTe and CeSbTe. In LaSbTe, the relaxation of quasiparticles is dominated by two different mechanisms: electron-phonon coupling, and phonon-assisted electron-hole recombination. Significantly, the amplitude of photoinduced reflectivity related to the former one shows two pronounced peaks at 156 K and 263 K, indicating the occurrence of two charge density wave (CDW) phase transitions. The ultrafast responses of CeSbTe share a lot of similarities with LaSbTe, and an additional CDW phase transition at 154 K is revealed in CeSbTe. However, the slower relaxation of CeSbTe exhibits an exotic behavior that deviates from the typical phonon-assisted electron-hole recombination process, probably due to the imbalance between the electron- and hole-type carriers. Unlike LaSbTe, the relaxation times of CeSbTe vary slightly with the pump power, inferring the possible participation of 4$f$ electron in the decay process. In addition, two oscillation modes around 1 THz and 3 THz are identified in both LaSbTe and CeSbTe, which are mostly likely to be coherent phonon modes. These findings unravel the existence of multiple CDW orders in LaSbTe and CeSbTe, offering insights into the underlying physics of these systems.

cond-mat.str-el↗

arXiv2024

Double fast algorithm for solving time-space fractional diffusion problems with spectral fractional Laplacian

This paper presents an efficient and concise double fast algorithm to solve high dimensional time-space fractional diffusion problems with spectral fractional Laplacian. We first establish semi-discrete scheme of time-space fractional diffusion equation, which uses linear finite element or fourth-order compact difference method combining with matrix transfer technique to approximate spectral fractional Laplacian. Then we introduce a fast time-stepping L1 scheme for time discretization. The proposed scheme can exactly evaluate fractional power of matrix and perform matrix-vector multiplication at per time level by using discrete sine transform, which doesn't need to resort to any iteration method and can significantly reduce computation cost and memory requirement. Further, we address stability and convergence analyses of full discrete scheme based on fast time-stepping L1 scheme on graded time mesh. Our analysis shows that the choice of graded mesh factor $ω=(2-α)/α$ shall give an optimal temporal convergence $\mathcal{O}(N^{-(2-α)})$ with $N$ denoting the number of time mesh. Finally, ample numerical examples are delivered to illustrate our theoretical analysis and the efficiency of the suggested scheme.

math.NA↗

arXiv2024

The Early Ultraviolet Light-Curves of Type II Supernovae and the Radii of Their Progenitor Stars

We present a sample of 34 normal SNe II detected with the Zwicky Transient Facility, with multi-band UV light-curves starting at $t \leq 4$ days after explosion, as well as X-ray detections and upper limits. We characterize the early UV-optical colors and provide prescriptions for empirical host-extinction corrections. We show that the $t > 2\,$days UV-optical colors and the blackbody evolution of the sample are consistent with the predictions of spherical phase shock-cooling (SC), independently of the presence of `flash ionization" features. We present a framework for fitting SC models which can reproduce the parameters of a set of multi-group simulations without a significant bias up to 20% in radius and velocity. Observations of about half of the SNe II in the sample are well-fit by models with breakout radii $<10^{14}\,$cm. The other half are typically more luminous, with observations from day 1 onward that are better fit by a model with a large $>10^{14}\,$cm breakout radius. However, these fits predict an early rise during the first day that is too slow. We suggest these large-breakout events are explosions of stars with an inflated envelope or a confined CSM with a steep density profile, at which breakout occurs. Using the X-ray data, we derive constraints on the extended ($\sim10^{15}$ cm) CSM density independent of spectral modeling, and find most SNe II progenitors lose $<10^{-4} M_{\odot}\, \rm yr^{-1}$ a few years before explosion. This provides independent evidence the CSM around many SNe II progenitors is confined. We show that the overall observed breakout radius distribution is skewed to higher radii due to a luminosity bias. We argue that the $66^{+11}_{-22}\%$ of red supergiants (RSG) explode as SNe II with breakout radii consistent with the observed distribution of field RSG, with a tail extending to large radii, likely due to the presence of CSM.

astro-ph.HE↗

arXiv2024

NICEST: Noisy Label Correction and Training for Robust Scene Graph Generation

Nearly all existing scene graph generation (SGG) models have overlooked the ground-truth annotation qualities of mainstream SGG datasets, i.e., they assume: 1) all the manually annotated positive samples are equally correct; 2) all the un-annotated negative samples are absolutely background. In this paper, we argue that neither of the assumptions applies to SGG: there are numerous noisy ground-truth predicate labels that break these two assumptions and harm the training of unbiased SGG models. To this end, we propose a novel NoIsy label CorrEction and Sample Training strategy for SGG: NICEST. Specifically, it consists of two parts: NICE and NIST, which rule out these noisy label issues by generating high-quality samples and the effective training strategy, respectively. NICE first detects noisy samples and then reassigns them more high-quality soft predicate labels. NIST is a multi-teacher knowledge distillation based training strategy, which enables the model to learn unbiased fusion knowledge. And a dynamic trade-off weighting strategy in NIST is designed to penalize the bias of different teachers. Due to the model-agnostic nature of both NICE and NIST, our NICEST can be seamlessly incorporated into any SGG architecture to boost its performance on different predicate categories. In addition, to better evaluate the generalization of SGG models, we further propose a new benchmark VG-OOD, by re-organizing the prevalent VG dataset and deliberately making the predicate distributions of the training and test sets as different as possible for each subject-object category pair. This new benchmark helps disentangle the influence of subject-object category based frequency biases. Extensive ablations and results on different backbones and tasks have attested to the effectiveness and generalization ability of each component of NICEST.

cs.CV↗

arXiv2024

Joint Conditional Diffusion Model for Image Restoration with Mixed Degradations

Image restoration is rather challenging in adverse weather conditions, especially when multiple degradations occur simultaneously. Blind image decomposition was proposed to tackle this issue, however, its effectiveness heavily relies on the accurate estimation of each component. Although diffusion-based models exhibit strong generative abilities in image restoration tasks, they may generate irrelevant contents when the degraded images are severely corrupted. To address these issues, we leverage physical constraints to guide the whole restoration process, where a mixed degradation model based on atmosphere scattering model is constructed. Then we formulate our Joint Conditional Diffusion Model (JCDM) by incorporating the degraded image and degradation mask to provide precise guidance. To achieve better color and detail recovery results, we further integrate a refinement network to reconstruct the restored image, where Uncertainty Estimation Block (UEB) is employed to enhance the features. Extensive experiments performed on both multi-weather and weather-specific datasets demonstrate the superiority of our method over state-of-the-art competing methods.

cs.CV↗

arXiv2024

SIFU: Side-view Conditioned Implicit Function for Real-world Usable Clothed Human Reconstruction

Creating high-quality 3D models of clothed humans from single images for real-world applications is crucial. Despite recent advancements, accurately reconstructing humans in complex poses or with loose clothing from in-the-wild images, along with predicting textures for unseen areas, remains a significant challenge. A key limitation of previous methods is their insufficient prior guidance in transitioning from 2D to 3D and in texture prediction. In response, we introduce SIFU (Side-view Conditioned Implicit Function for Real-world Usable Clothed Human Reconstruction), a novel approach combining a Side-view Decoupling Transformer with a 3D Consistent Texture Refinement pipeline.SIFU employs a cross-attention mechanism within the transformer, using SMPL-X normals as queries to effectively decouple side-view features in the process of mapping 2D features to 3D. This method not only improves the precision of the 3D models but also their robustness, especially when SMPL-X estimates are not perfect. Our texture refinement process leverages text-to-image diffusion-based prior to generate realistic and consistent textures for invisible views. Through extensive experiments, SIFU surpasses SOTA methods in both geometry and texture reconstruction, showcasing enhanced robustness in complex scenarios and achieving an unprecedented Chamfer and P2S measurement. Our approach extends to practical applications such as 3D printing and scene building, demonstrating its broad utility in real-world scenarios. Project page https://river-zhang.github.io/SIFU-projectpage/ .

cs.CV↗

arXiv2024

LGSDF: Continual Global Learning of Signed Distance Fields Aided by Local Updating

Implicit reconstruction of ESDF (Euclidean Signed Distance Field) involves training a neural network to regress the signed distance from any point to the nearest obstacle, which has the advantages of lightweight storage and continuous querying. However, existing algorithms usually rely on conflicting raw observations as training data, resulting in poor map performance. In this paper, we propose LGSDF, an ESDF continual Global learning algorithm aided by Local updating. At the front end, axis-aligned grids are dynamically updated by pre-processed sensor observations, where incremental fusion alleviates estimation error caused by limited viewing directions. At the back end, a randomly initialized implicit ESDF neural network performs continual self-supervised learning guided by these grids to generate smooth and continuous maps. The results on multiple scenes show that LGSDF can construct more accurate ESDF maps and meshes compared with SOTA (State Of The Art) explicit and implicit mapping algorithms. The source code of LGSDF is publicly available at https://github.com/BIT-DYN/LGSDF.

cs.CV↗

arXiv2024

StepNet: Spatial-temporal Part-aware Network for Isolated Sign Language Recognition

The goal of sign language recognition (SLR) is to help those who are hard of hearing or deaf overcome the communication barrier. Most existing approaches can be typically divided into two lines, i.e., Skeleton-based and RGB-based methods, but both the two lines of methods have their limitations. Skeleton-based methods do not consider facial expressions, while RGB-based approaches usually ignore the fine-grained hand structure. To overcome both limitations, we propose a new framework called Spatial-temporal Part-aware network~(StepNet), based on RGB parts. As its name suggests, it is made up of two modules: Part-level Spatial Modeling and Part-level Temporal Modeling. Part-level Spatial Modeling, in particular, automatically captures the appearance-based properties, such as hands and faces, in the feature space without the use of any keypoint-level annotations. On the other hand, Part-level Temporal Modeling implicitly mines the long-short term context to capture the relevant attributes over time. Extensive experiments demonstrate that our StepNet, thanks to spatial-temporal modules, achieves competitive Top-1 Per-instance accuracy on three commonly-used SLR benchmarks, i.e., 56.89% on WLASL, 77.2% on NMFs-CSL, and 77.1% on BOBSL. Additionally, the proposed method is compatible with the optical flow input and can produce superior performance if fused. For those who are hard of hearing, we hope that our work can act as a preliminary step.

cs.CV↗

arXiv2024

Visual Knowledge in the Big Model Era: Retrospect and Prospect

Visual knowledge is a new form of knowledge representation that can encapsulate visual concepts and their relations in a succinct, comprehensive, and interpretable manner, with a deep root in cognitive psychology. As the knowledge about the visual world has been identified as an indispensable component of human cognition and intelligence, visual knowledge is poised to have a pivotal role in establishing machine intelligence. With the recent advance of Artificial Intelligence (AI) techniques, large AI models (or foundation models) have emerged as a potent tool capable of extracting versatile patterns from broad data as implicit knowledge, and abstracting them into an outrageous amount of numeric parameters. To pave the way for creating visual knowledge empowered AI machines in this coming wave, we present a timely review that investigates the origins and development of visual knowledge in the pre-big model era, and accentuates the opportunities and unique role of visual knowledge in the big model era.

cs.AI↗

arXiv2024

Meta Invariance Defense Towards Generalizable Robustness to Unknown Adversarial Attacks

Despite providing high-performance solutions for computer vision tasks, the deep neural network (DNN) model has been proved to be extremely vulnerable to adversarial attacks. Current defense mainly focuses on the known attacks, but the adversarial robustness to the unknown attacks is seriously overlooked. Besides, commonly used adaptive learning and fine-tuning technique is unsuitable for adversarial defense since it is essentially a zero-shot problem when deployed. Thus, to tackle this challenge, we propose an attack-agnostic defense method named Meta Invariance Defense (MID). Specifically, various combinations of adversarial attacks are randomly sampled from a manually constructed Attacker Pool to constitute different defense tasks against unknown attacks, in which a student encoder is supervised by multi-consistency distillation to learn the attack-invariant features via a meta principle. The proposed MID has two merits: 1) Full distillation from pixel-, feature- and prediction-level between benign and adversarial samples facilitates the discovery of attack-invariance. 2) The model simultaneously achieves robustness to the imperceptible adversarial perturbations in high-level image classification and attack-suppression in low-level robust image regeneration. Theoretical and empirical studies on numerous benchmarks such as ImageNet verify the generalizable robustness and superiority of MID under various attacks.

cs.CV↗

arXiv2024

Maximal quantum interaction between free electrons and photons

The emerging field of free-electron quantum optics enables electron-photon entanglement and holds the potential for generating nontrivial photon states for quantum information processing. Although recent experimental studies have entered the quantum regime, rapid theoretical developments predict that qualitatively unique phenomena only emerge beyond a certain interaction strength. It is thus pertinent to identify the maximal electron-photon interaction strength and the materials, geometries, and particle energies that enable one to approach it. We derive an upper limit to the quantum vacuum interaction strength between free electrons and single-mode photons, which illuminates the conditions for the strongest interaction. Crucially, we obtain an explicit energy selection recipe for electrons and photons to achieve maximal interaction at arbitrary separations and identify two optimal regimes favoring either fast or slow electrons over those with intermediate velocities. We validate the limit by analytical and numerical calculations on canonical geometries and provide near-optimal designs indicating the feasibility of strong quantum interactions. Our findings offer fundamental intuition for maximizing the quantum interaction between free electrons and photons and provide practical design rules for future experiments on electron-photon and electron-mediated photon-photon entanglement. They should also enable the evaluation of key metrics for applications such as the maximum power of free-electron radiation sources and the maximum acceleration gradient of dielectric laser accelerators.

quant-ph↗

arXiv2024

Improving Bird's Eye View Semantic Segmentation by Task Decomposition

Semantic segmentation in bird's eye view (BEV) plays a crucial role in autonomous driving. Previous methods usually follow an end-to-end pipeline, directly predicting the BEV segmentation map from monocular RGB inputs. However, the challenge arises when the RGB inputs and BEV targets from distinct perspectives, making the direct point-to-point predicting hard to optimize. In this paper, we decompose the original BEV segmentation task into two stages, namely BEV map reconstruction and RGB-BEV feature alignment. In the first stage, we train a BEV autoencoder to reconstruct the BEV segmentation maps given corrupted noisy latent representation, which urges the decoder to learn fundamental knowledge of typical BEV patterns. The second stage involves mapping RGB input images into the BEV latent space of the first stage, directly optimizing the correlations between the two views at the feature level. Our approach simplifies the complexity of combining perception and generation into distinct steps, equipping the model to handle intricate and challenging scenes effectively. Besides, we propose to transform the BEV segmentation map from the Cartesian to the polar coordinate system to establish the column-wise correspondence between RGB images and BEV maps. Moreover, our method requires neither multi-scale features nor camera intrinsic parameters for depth estimation and saves computational overhead. Extensive experiments on nuScenes and Argoverse show the effectiveness and efficiency of our method. Code is available at https://github.com/happytianhao/TaDe.

cs.CV↗

arXiv2024

A Shock Flash Breaking Out of a Dusty Red Supergiant

Shock breakout emission is light that arises when a shockwave, generated by core-collapse explosion of a massive star, passes through its outer envelope. Hitherto, the earliest detection of such a signal was at several hours after the explosion, though a few others had been reported. The temporal evolution of early light curves should reveal insights into the shock propagation, including explosion asymmetry and environment in the vicinity, but this has been hampered by the lack of multiwavelength observations. Here we report the instant multiband observations of a type II supernova (SN 2023ixf) in the galaxy M101 (at a distance of 6.85+/-0.15 Mpc), beginning at about 1.4 hours after the explosion. The exploding star was a red supergiant with a radius of about 440 solar radii. The light curves evolved rapidly, on timescales of 1-2 hours, and appeared unusually fainter and redder than predicted by models within the first few hours, which we attribute to an optically thick dust shell before it was disrupted by the shockwave. We infer that the breakout and perhaps the distribution of the surrounding dust were not spherically symmetric.

astro-ph.HE↗

arXiv2024

Clustering for Protein Representation Learning

Protein representation learning is a challenging task that aims to capture the structure and function of proteins from their amino acid sequences. Previous methods largely ignored the fact that not all amino acids are equally important for protein folding and activity. In this article, we propose a neural clustering framework that can automatically discover the critical components of a protein by considering both its primary and tertiary structure information. Our framework treats a protein as a graph, where each node represents an amino acid and each edge represents a spatial or sequential connection between amino acids. We then apply an iterative clustering strategy to group the nodes into clusters based on their 1D and 3D positions and assign scores to each cluster. We select the highest-scoring clusters and use their medoid nodes for the next iteration of clustering, until we obtain a hierarchical and informative representation of the protein. We evaluate on four protein-related tasks: protein fold classification, enzyme reaction classification, gene ontology term prediction, and enzyme commission number prediction. Experimental results demonstrate that our method achieves state-of-the-art performance.

cs.LG↗

arXiv2024

GD^2-NeRF: Generative Detail Compensation via GAN and Diffusion for One-shot Generalizable Neural Radiance Fields

In this paper, we focus on the One-shot Novel View Synthesis (O-NVS) task which targets synthesizing photo-realistic novel views given only one reference image per scene. Previous One-shot Generalizable Neural Radiance Fields (OG-NeRF) methods solve this task in an inference-time finetuning-free manner, yet suffer the blurry issue due to the encoder-only architecture that highly relies on the limited reference image. On the other hand, recent diffusion-based image-to-3d methods show vivid plausible results via distilling pre-trained 2D diffusion models into a 3D representation, yet require tedious per-scene optimization. Targeting these issues, we propose the GD$^2$-NeRF, a Generative Detail compensation framework via GAN and Diffusion that is both inference-time finetuning-free and with vivid plausible details. In detail, following a coarse-to-fine strategy, GD$^2$-NeRF is mainly composed of a One-stage Parallel Pipeline (OPP) and a 3D-consistent Detail Enhancer (Diff3DE). At the coarse stage, OPP first efficiently inserts the GAN model into the existing OG-NeRF pipeline for primarily relieving the blurry issue with in-distribution priors captured from the training dataset, achieving a good balance between sharpness (LPIPS, FID) and fidelity (PSNR, SSIM). Then, at the fine stage, Diff3DE further leverages the pre-trained image diffusion models to complement rich out-distribution details while maintaining decent 3D consistency. Extensive experiments on both the synthetic and real-world datasets show that GD$^2$-NeRF noticeably improves the details while without per-scene finetuning.

cs.CV↗

arXiv2024

Psychometry: An Omnifit Model for Image Reconstruction from Human Brain Activity

Reconstructing the viewed images from human brain activity bridges human and computer vision through the Brain-Computer Interface. The inherent variability in brain function between individuals leads existing literature to focus on acquiring separate models for each individual using their respective brain signal data, ignoring commonalities between these data. In this article, we devise Psychometry, an omnifit model for reconstructing images from functional Magnetic Resonance Imaging (fMRI) obtained from different subjects. Psychometry incorporates an omni mixture-of-experts (Omni MoE) module where all the experts work together to capture the inter-subject commonalities, while each expert associated with subject-specific parameters copes with the individual differences. Moreover, Psychometry is equipped with a retrieval-enhanced inference strategy, termed Ecphory, which aims to enhance the learned fMRI representation via retrieving from prestored subject-specific memories. These designs collectively render Psychometry omnifit and efficient, enabling it to capture both inter-subject commonality and individual specificity across subjects. As a result, the enhanced fMRI representations serve as conditional signals to guide a generation model to reconstruct high-quality and realistic images, establishing Psychometry as state-of-the-art in terms of both high-level and low-level metrics.

cs.CV↗