arXiv subjects

Hong Liu

Publications and source records attributed to Hong Liu.

At least 181 records · Page 10Linked to original sources

arXiv2024

Chaos and the Emergence of the Cosmological Horizon

We construct algebras of diff-invariant observables in a global de Sitter universe with two observers and a free scalar QFT in two dimensions. We work in the strict $G_N \rightarrow 0$ limit, but allow the observers to have an order one mass in cosmic units. The observers are fully quantized. In the limit when the observers have infinite mass and are localized along geodesics at the North and South poles, it was shown in previous work \cite{CLPW} that their algebras are mutually commuting type II$_1$ factors. Away from this limit, we show that the algebras fail to commute and that they are type I non-factors. Physically, this is because the observers' trajectories are uncertain and state-dependent, and they may come into causal contact. We compute out-of-time-ordered correlators along an observer's worldline, and observe a Lyapunov exponent given by $\frac{4 π}{β_{\text{dS}}}$, as a result of observer recoil and de Sitter expansion. This should be contrasted with results from AdS gravity, and exceeds the chaos bound associated with the de Sitter temperature by a factor of two. We also discuss how the cosmological horizon emerges in the large mass limit and comment on implications for de Sitter holography.

hep-th↗

arXiv2024

D$^3$epth: Self-Supervised Depth Estimation with Dynamic Mask in Dynamic Scenes

Depth estimation is a crucial technology in robotics. Recently, self-supervised depth estimation methods have demonstrated great potential as they can efficiently leverage large amounts of unlabelled real-world data. However, most existing methods are designed under the assumption of static scenes, which hinders their adaptability in dynamic environments. To address this issue, we present D$^3$epth, a novel method for self-supervised depth estimation in dynamic scenes. It tackles the challenge of dynamic objects from two key perspectives. First, within the self-supervised framework, we design a reprojection constraint to identify regions likely to contain dynamic objects, allowing the construction of a dynamic mask that mitigates their impact at the loss level. Second, for multi-frame depth estimation, we introduce a cost volume auto-masking strategy that leverages adjacent frames to identify regions associated with dynamic objects and generate corresponding masks. This provides guidance for subsequent processes. Furthermore, we propose a spectral entropy uncertainty module that incorporates spectral entropy to guide uncertainty estimation during depth fusion, effectively addressing issues arising from cost volume computation in dynamic environments. Extensive experiments on KITTI and Cityscapes datasets demonstrate that the proposed method consistently outperforms existing self-supervised monocular depth estimation baselines. Code is available at \url{https://github.com/Csyunling/D3epth}.

cs.CV↗

arXiv2024

Engineered Dual BIC Resonances in Hybrid Metasurfaces for Controlled Photoluminescence Amplification

The development of miniaturized light sources with tunable functionality is crucial for advancing integrated photonic devices, enabling applications in quantum computing, communications, and sensing. Achieving tunable light emission after device fabrication remains a significant challenge, particularly when efficient amplification is required. Hybrid metasurfaces, which integrate several nanostructured materials to form optical resonators, have emerged as promising candidates to overcome these limitations by providing a high degree of flexibility in emission control and enhanced amplification. In this work, we demonstrate tunable amplified photoluminescence (PL) in nanocrystalline silicon (nc-Si) quantum dots (QDs) embedded in a hybrid metasurface consisting of amorphous silicon (a-Si) and a low-loss phase change material (PCM) antimony trisulfide (Sb2S3). The nc-Si QDs maintain a high PL efficiency and stability at elevated temperatures, offering reliable and tunable phase transitions in the PCM. The hybrid metasurface supports dual quasi-bound states in the continuum (BICs) to achieve Q-factors up to 225. The dual BIC cavity enables tunable amplified PL by a factor of 15 with a wavelength shift of up to 105 nm via dimensional modulation. Meanwhile, all-optical tunable PL emission across a 24 nm wavelength range has been achieved when PCMs are tuned from the amorphous to crystalline phase. Furthermore, we propose a high Q-factor metalens to focus the tunable amplified PL, extending the diffraction-limited focusing tunability into the near infrared (NIR). This work paves the way for highly efficient quantum light sources using reconfigurable nanophotonic devices in next-generation photonic systems.

physics.optics↗

arXiv2024

SCMIL: Sparse Context-aware Multiple Instance Learning for Predicting Cancer Survival Probability Distribution in Whole Slide Images

Cancer survival prediction is a challenging task that involves analyzing of the tumor microenvironment within Whole Slide Image (WSI). Previous methods cannot effectively capture the intricate interaction features among instances within the local area of WSI. Moreover, existing methods for cancer survival prediction based on WSI often fail to provide better clinically meaningful predictions. To overcome these challenges, we propose a Sparse Context-aware Multiple Instance Learning (SCMIL) framework for predicting cancer survival probability distributions. SCMIL innovatively segments patches into various clusters based on their morphological features and spatial location information, subsequently leveraging sparse self-attention to discern the relationships between these patches with a context-aware perspective. Considering many patches are irrelevant to the task, we introduce a learnable patch filtering module called SoftFilter, which ensures that only interactions between task-relevant patches are considered. To enhance the clinical relevance of our prediction, we propose a register-based mixture density network to forecast the survival probability distribution for individual patients. We evaluate SCMIL on two public WSI datasets from the The Cancer Genome Atlas (TCGA) specifically focusing on lung adenocarcinom (LUAD) and kidney renal clear cell carcinoma (KIRC). Our experimental results indicate that SCMIL outperforms current state-of-the-art methods for survival prediction, offering more clinically meaningful and interpretable outcomes. Our code is accessible at https://github.com/yang-ze-kang/SCMIL.

cs.CV↗

arXiv2024

ARTS: Semi-Analytical Regressor using Disentangled Skeletal Representations for Human Mesh Recovery from Videos

Although existing video-based 3D human mesh recovery methods have made significant progress, simultaneously estimating human pose and shape from low-resolution image features limits their performance. These image features lack sufficient spatial information about the human body and contain various noises (e.g., background, lighting, and clothing), which often results in inaccurate pose and inconsistent motion. Inspired by the rapid advance in human pose estimation, we discover that compared to image features, skeletons inherently contain accurate human pose and motion. Therefore, we propose a novel semiAnalytical Regressor using disenTangled Skeletal representations for human mesh recovery from videos, called ARTS. Specifically, a skeleton estimation and disentanglement module is proposed to estimate the 3D skeletons from a video and decouple them into disentangled skeletal representations (i.e., joint position, bone length, and human motion). Then, to fully utilize these representations, we introduce a semi-analytical regressor to estimate the parameters of the human mesh model. The regressor consists of three modules: Temporal Inverse Kinematics (TIK), Bone-guided Shape Fitting (BSF), and Motion-Centric Refinement (MCR). TIK utilizes joint position to estimate initial pose parameters and BSF leverages bone length to regress bone-aligned shape parameters. Finally, MCR combines human motion representation with image features to refine the initial human model parameters. Extensive experiments demonstrate that our ARTS surpasses existing state-of-the-art video-based methods in both per-frame accuracy and temporal consistency on popular benchmarks: 3DPW, MPI-INF-3DHP, and Human3.6M. Code is available at https://github.com/TangTao-PKU/ARTS.

cs.CV↗

arXiv2024

Using Zone Inflation and Volume Transfer to Design a Fabric-based Pneumatic Exosuit with both Efficiency and Wearability

Fabric-based pneumatic exosuits have a broad application prospect due to their good human-machine interaction performance, but their structural design paradigm has not yet been finalized and requires in-depth research. This paper proposes the concepts of zone inflation and volume transfer for the design of a fabric-based pneumatic exosuit with both efficiency and wearability. The meaning of zone inflation is to divide the inflation area of pneumatic exosuit into inflation-deflation zone and inflation-holding zone which can reduce the consumption of compressed air and improve efficiency. Volume transfer, a strategic distribution method of inflatable regions inside the garment, can effectively enhance the wearability of the exosuit. Using inexpensive thermoplastic polyurethane film and clothing fabric, the exosuit is made by heat pressing and sewing. The exosuit has a response time of 0.5s, a stress area of 1500mm2, and a profile of only 32mm, which can be hidden inside common clothing. A mathematical model is developed to predict the output torque of the exosuit with an error of 3.6%. Mechanical experiments show that the exosuit outputs a torque of 9.1Nm at a pressure of 100kPa. Surface electromyography experiments show that the exosuit can provide users with a boost from sitting to standing, with an average reduction in electromyography signals of 14.95%. The exosuit designed using these methods synthesizes efficiency and wearability and is expected to be an ideal paradigm for fabric-based pneumatic exosuits.

cs.RO↗

arXiv2024

Clique density vs blowups

A well-known theorem of Nikiforov asserts that any graph with a positive $K_{r}$-density contains a logarithmic blowup of $K_r$. In this paper, we explore variants of Nikiforov's result in the following form. Given $r,t\in\mathbb{N}$, when a positive $K_{r}$-density implies the existence of a significantly larger (with almost linear size) blowup of $K_t$? Our results include: For an $n$-vertex ordered graph $G$ with no induced monotone path $P_{6}$, if its complement $\overline{G}$ has positive triangle density, then $\overline{G}$ contains a biclique of size $Ω(\frac{n}{\log{n}})$. This strengthens a recent result of Pach and Tomon. For general $k$, let $g(k)$ be the minimum $r\in \mathbb{N}$ such that for any $n$-vertex ordered graph $G$ with no induced monotone $P_{2k}$, if $\overline{G}$ has positive $K_r$-density, then $\overline{G}$ contains a biclique of size $Ω(\frac{n}{\log{n}})$. Using concentration of measure and the isodiametric inequality on high dimensional spheres, we provide constructions showing that, surprisingly, $g(k)$ grows quadratically. On the other hand, we relate the problem of upper bounding $g(k)$ to a certain Ramsey problem and determine $g(k)$ up to a factor of 2. Any incomparability graph with positive $K_{r}$-density contains a blowup of $K_r$ of size $Ω(\frac{n}{\log{n}}).$ This confirms a conjecture of Tomon in a stronger form. In doing so, we obtain a strong regularity type lemma for incomparability graphs with no large blowups of a clique, which is of independent interest. We also prove that any $r$-comparability graph with positive $K_{(2h-2)^{r}+1}$-density contains a blowup of $K_h$ of size $Ω(n)$, where the constant $(2h-2)^{r}+1$ is optimal. The $\frac{n}{\log n}$ size of the blowups in all our results are optimal up to a constant factor.

math.CO↗

arXiv2024

STNet: Deep Audio-Visual Fusion Network for Robust Speaker Tracking

Audio-visual speaker tracking aims to determine the location of human targets in a scene using signals captured by a multi-sensor platform, whose accuracy and robustness can be improved by multi-modal fusion methods. Recently, several fusion methods have been proposed to model the correlation in multiple modalities. However, for the speaker tracking problem, the cross-modal interaction between audio and visual signals hasn't been well exploited. To this end, we present a novel Speaker Tracking Network (STNet) with a deep audio-visual fusion model in this work. We design a visual-guided acoustic measurement method to fuse heterogeneous cues in a unified localization space, which employs visual observations via a camera model to construct the enhanced acoustic map. For feature fusion, a cross-modal attention module is adopted to jointly model multi-modal contexts and interactions. The correlated information between audio and visual features is further interacted in the fusion model. Moreover, the STNet-based tracker is applied to multi-speaker cases by a quality-aware module, which evaluates the reliability of multi-modal observations to achieve robust tracking in complex scenarios. Experiments on the AV16.3 and CAV3D datasets show that the proposed STNet-based tracker outperforms uni-modal methods and state-of-the-art audio-visual speaker trackers.

cs.CV↗

arXiv2024

Strong lensing as a probe of braneworld

For the first time, we use the Event Horizon Telescope (EHT) data to constrain the parameters of braneworld black holes which constrain $ε=0.0285^{+0.0888+0.1456}_{-0.0895-0.1475}$ for the anisotropic black hole and $q=-0.0305^{+0.1034+0.1953}_{-0.0895-0.1470}$ for the tidal Reissner-Nordstr$\ddot{o}$m (RN) black hole. Based on the fitted data and physical requirement, we calculate the photon deflection, the angular separation and time delay between different relativistic images of the anisotropic black hole and the tidal RN black hole in the ranges $-0.1190<ε<0$ and $-0.1775 <q<0$. And furthermore, we study the quasinormal modes (QNMs) for the braneworld black holes. The results shed light on existence of extra dimension.

gr-qc↗

arXiv2024

DiffusionFake: Enhancing Generalization in Deepfake Detection via Guided Stable Diffusion

The rapid progress of Deepfake technology has made face swapping highly realistic, raising concerns about the malicious use of fabricated facial content. Existing methods often struggle to generalize to unseen domains due to the diverse nature of facial manipulations. In this paper, we revisit the generation process and identify a universal principle: Deepfake images inherently contain information from both source and target identities, while genuine faces maintain a consistent identity. Building upon this insight, we introduce DiffusionFake, a novel plug-and-play framework that reverses the generative process of face forgeries to enhance the generalization of detection models. DiffusionFake achieves this by injecting the features extracted by the detection model into a frozen pre-trained Stable Diffusion model, compelling it to reconstruct the corresponding target and source images. This guided reconstruction process constrains the detection network to capture the source and target related features to facilitate the reconstruction, thereby learning rich and disentangled representations that are more resilient to unseen forgeries. Extensive experiments demonstrate that DiffusionFake significantly improves cross-domain generalization of various detector architectures without introducing additional parameters during inference. Our Codes are available in https://github.com/skJack/DiffusionFake.git.

cs.CV↗

arXiv2024

Cross-Model Cross-Stream Learning for Self-Supervised Human Action Recognition

Considering the instance-level discriminative ability, contrastive learning methods, including MoCo and SimCLR, have been adapted from the original image representation learning task to solve the self-supervised skeleton-based action recognition task. These methods usually use multiple data streams (i.e., joint, motion, and bone) for ensemble learning, meanwhile, how to construct a discriminative feature space within a single stream and effectively aggregate the information from multiple streams remains an open problem. To this end, this paper first applies a new contrastive learning method called BYOL to learn from skeleton data, and then formulate SkeletonBYOL as a simple yet effective baseline for self-supervised skeleton-based action recognition. Inspired by SkeletonBYOL, this paper further presents a Cross-Model and Cross-Stream (CMCS) framework. This framework combines Cross-Model Adversarial Learning (CMAL) and Cross-Stream Collaborative Learning (CSCL). Specifically, CMAL learns single-stream representation by cross-model adversarial loss to obtain more discriminative features. To aggregate and interact with multi-stream information, CSCL is designed by generating similarity pseudo label of ensemble learning as supervision and guiding feature generation for individual streams. Extensive experiments on three datasets verify the complementary properties between CMAL and CSCL and also verify that the proposed method can achieve better results than state-of-the-art methods using various evaluation protocols.

cs.CV↗

arXiv2024

New lower bound on ball packing density in high-dimensional hyperbolic spaces

We present a new lower bound on the Bowen-Radin maximal density of radius-R ball packings in the m-dimensional hyperbolic space, improving on the basic covering bound by factor Ω(m(R+\ln m)) as m tends to infinity. This is done by applying the recent theorem of Campos, Jenssen, Michelen and Sahasrabudhe on independent sets in graphs with sparse neighbourhoods.

math.CO↗

arXiv2024

GraphMLP: A Graph MLP-Like Architecture for 3D Human Pose Estimation

Modern multi-layer perceptron (MLP) models have shown competitive results in learning visual representations without self-attention. However, existing MLP models are not good at capturing local details and lack prior knowledge of human body configurations, which limits their modeling power for skeletal representation learning. To address these issues, we propose a simple yet effective graph-reinforced MLP-Like architecture, named GraphMLP, that combines MLPs and graph convolutional networks (GCNs) in a global-local-graphical unified architecture for 3D human pose estimation. GraphMLP incorporates the graph structure of human bodies into an MLP model to meet the domain-specific demand of the 3D human pose, while allowing for both local and global spatial interactions. Furthermore, we propose to flexibly and efficiently extend the GraphMLP to the video domain and show that complex temporal dynamics can be effectively modeled in a simple way with negligible computational cost gains in the sequence length. To the best of our knowledge, this is the first MLP-Like architecture for 3D human pose estimation in a single frame and a video sequence. Extensive experiments show that the proposed GraphMLP achieves state-of-the-art performance on two datasets, i.e., Human3.6M and MPI-INF-3DHP. Code and models are available at https://github.com/Vegetebird/GraphMLP.

cs.CV↗

arXiv2024

Chain of Thought Empowers Transformers to Solve Inherently Serial Problems

Instructing the model to generate a sequence of intermediate steps, a.k.a., a chain of thought (CoT), is a highly effective method to improve the accuracy of large language models (LLMs) on arithmetics and symbolic reasoning tasks. However, the mechanism behind CoT remains unclear. This work provides a theoretical understanding of the power of CoT for decoder-only transformers through the lens of expressiveness. Conceptually, CoT empowers the model with the ability to perform inherently serial computation, which is otherwise lacking in transformers, especially when depth is low. Given input length $n$, previous works have shown that constant-depth transformers with finite precision $\mathsf{poly}(n)$ embedding size can only solve problems in $\mathsf{TC}^0$ without CoT. We first show an even tighter expressiveness upper bound for constant-depth transformers with constant-bit precision, which can only solve problems in $\mathsf{AC}^0$, a proper subset of $ \mathsf{TC}^0$. However, with $T$ steps of CoT, constant-depth transformers using constant-bit precision and $O(\log n)$ embedding size can solve any problem solvable by boolean circuits of size $T$. Empirically, enabling CoT dramatically improves the accuracy for tasks that are hard for parallel computation, including the composition of permutation groups, iterated squaring, and circuit value problems, especially for low-depth transformers.

cs.LG↗

arXiv2024

New horizon symmetries, hydrodynamics, and quantum chaos

We generalize the formulation of horizon symmetries presented in previous literature to include diffeomorphisms that can shift the location of the horizon. In the context of the AdS/CFT duality, we show that horizon symmetries can be interpreted on the boundary as emergent low-energy gauge symmetries. In particular, we identify a new class of horizon symmetries that extend the so-called shift symmetry, which was previously postulated for effective field theories of maximally chaotic systems. Additionally, we comment on the connections of horizon symmetries with bulk calculations of out-of-time-ordered correlation functions and the phenomenon of pole-skipping.

hep-th↗

arXiv2024

A Multimodal Object-level Contrast Learning Method for Cancer Survival Risk Prediction

Computer-aided cancer survival risk prediction plays an important role in the timely treatment of patients. This is a challenging weakly supervised ordinal regression task associated with multiple clinical factors involved such as pathological images, genomic data and etc. In this paper, we propose a new training method, multimodal object-level contrast learning, for cancer survival risk prediction. First, we construct contrast learning pairs based on the survival risk relationship among the samples in the training sample set. Then we introduce the object-level contrast learning method to train the survival risk predictor. We further extend it to the multimodal scenario by applying cross-modal constrast. Considering the heterogeneity of pathological images and genomics data, we construct a multimodal survival risk predictor employing attention-based and self-normalizing based nerural network respectively. Finally, the survival risk predictor trained by our proposed method outperforms state-of-the-art methods on two public multimodal cancer datasets for survival risk prediction.

cs.LG↗

arXiv2024

Sidorenko's conjecture for subdivisions and theta substitutions

The famous Sidorenko's conjecture asserts that for every bipartite graph $H$, the number of homomorphisms from $H$ to a graph $G$ with given edge density is minimized when $G$ is pseudorandom. We prove that for any graph $H$, a graph obtained from replacing edges of $H$ by generalized theta graphs consisting of even paths satisfies Sidorenko's conjecture, provided a certain divisibility condition on the number of paths. To achieve this, we prove unconditionally that bipartite graphs obtained from replacing each edge of a complete graph with a generalized theta graph satisfy Sidorenko's conjecture, which extends a result of Conlon, Kim, Lee and Lee [J. Lond. Math. Soc., 2018].

math.CO↗

arXiv2024

Beyond chromatic threshold via the $(p,q)$-theorem, and a sharp blow-up phenomenon

We establish a novel connection between the well-known chromatic threshold problem in extremal combinatorics and the celebrated $(p,q)$-theorem in discrete geometry. In particular, for a graph $G$ with bounded clique number and a natural density condition, we prove a $(p,q)$-theorem for an abstract convexity space associated with $G$. Our result strengthens those of Thomassen and Nikiforov on the chromatic threshold of cliques. Our $(p,q)$-theorem can also be viewed as a $χ$-boundedness result for (what we call) ultra maximal $K_r$-free graphs. We further show that the graphs under study are blow-ups of constant size graphs, improving a result of Oberkampf and Schacht on homomorphism threshold of cliques. Our result unravels the cause underpinning such a blow-up phenomenon, differentiating the chromatic and homomorphism threshold problems for cliques. It implies that for the homomorphism threshold problem, rather than the minimum degree condition usually considered in the literature, the decisive factor is a clique density condition on co-neighborhoods of vertices. More precisely, we show that if an $n$-vertex $K_{r}$-free graph $G$ satisfies that the common neighborhood of every pair of non-adjacent vertices induces a subgraph with $K_{r-2}$-density at least $\varepsilon>0$, then $G$ must be a blow-up of some $K_r$-free graph $F$ on at most $2^{O(\frac{r}{\varepsilon}\log\frac{1}{\varepsilon})}$ vertices. Furthermore, this single exponential bound is optimal. We construct examples with no $K_r$-free homomorphic image of size smaller than $2^{Ω_r(\frac{1}{\varepsilon})}$.

math.CO↗