Searcharxiv⌕ Search

arXiv subjects

Jiahao Guo

Publications and source records attributed to Jiahao Guo.

14 recordsLinked to original sources

New Historical Guest-Star Candidates from Reassessed Chinese Records: Lunar Occultation of Planets

Ancient Chinese records of lunar occultations of planets have been used to test planetary theories and investigate long-term changes in Earth's rotation. However, certain records contain identification errors whose origins remain poorly understood; we propose that some of these errors may arise from misidentifying guest stars, such as novae or supernovae, as planets. We analyzed 67 errata on lunar occultations of planets from Liu Ci-Yuan's (2015) monograph using modern ephemerides and the open-source Python package \texttt{occultsky}, publicly available on GitHub. Using recorded dates and dynastic capital coordinates, we constructed observing windows anchored at solar noon and extending by $\pm 1$ day. Each window covered the preceding and recorded nights because the date may be off by one day. For each of the 67 records, we cataloged the apparent right ascension and declination $(α, δ)$ at the midpoint of the Moon's visibility window on each candidate night. The band swept by the lunar disk confines a possible transient to a narrow region a few degrees across; these midpoint coordinates serve as proxies for the implied guest-star positions. The catalog supplies candidate positions rather than identifications, enabling remnant and compact-object searches to test the guest-star hypothesis directly. If associations are established, these records could become dated guest-star candidates, and their dates could anchor remnant ages.

astro-ph.HE↗

MoETok: Sparse Expert Routing for 1D Visual Tokenizers

Modern visual tokenizers aim to preserve fine-grained visual details while learning high-level semantic representations for downstream generation and understanding. Query-based 1D visual tokenizers encode images into flexible latent sequences, yet how individual learnable queries gather and encode image information remains poorly understood. We reveal that these queries spontaneously develop complementary roles under semantic supervision: a small set of Content queries captures high-level semantic information, while a larger set of Detail queries preserves fine-grained visual details for reconstruction. To adapt token processing to these heterogeneous roles, we introduce MoETok, a 1D visual tokenizer that routes tokens through sparsely activated experts. Jointly optimized with reconstruction and semantic objectives, the router learns role-specialized processing paths adapted to each token's representation. Across model scales and token budgets, MoETok consistently improves reconstruction, semantic representation, and autoregressive generation over dense tokenizers under matched supervision and active computation. On ImageNet, it combines strong reconstruction and semantic representation with state-of-the-art class-conditional generation within the 1D tokenizer family. We further validate MoETok at scale in a unified autoregressive model jointly trained for understanding and generation with a next-token cross-entropy objective. The resulting model combines competitive visual understanding with state-of-the-art generation among unified discrete models, matching or surpassing diffusion-based models such as BAGEL and Qwen-Image on benchmarks including DPG-Bench (87.61), GenEval2 (40.04 GM), and WISE (0.49).

cs.CV↗

Redundant or Necessary? A Benchmark for Detecting Redundant Steps in Agent Trajectories

Computation cost is a critical challenge for LLM agent systems, as their long-horizon trajectories continuously accumulate computational overhead. While existing studies focus on improving the overall efficiency of agent systems, the efficiency of individual steps remains largely unexplored. In this paper, we propose and formulate a new research problem: \textbf{redundant step detection}, aiming to detect these steps that consume substantial computational resources while contributing little or nothing to task completion. To support this initiative, we introduce \textbf{RedundancyBench}, a new benchmark that contains 1,920 trajectories collected from 6 different LLMs with 3,613 step-level annotations. Using RedundancyBench, we develop and evaluate 6 frontier LLMs and Jev as redundant step detectors, summarizing their strengths and limitations. Our experiments reveal that the redundant step detection is still challenging even for SOTA models, such as Claude-Opus-5, particularly without prior information. Code and dataset in this paper are both available in https://anonymous.4open.science/r/RedundancyBench-E0CB.

cs.AI↗

XS-VID: A Large-Scale Benchmark for Small Object Detection and Tracking in Videos

Small object detection and tracking in videos remain critical yet underexplored challenges in computer vision, particularly for applications such as public safety, aerial surveillance, and autonomous driving. Existing benchmarks offer limited support due to limited numbers of small objects, constrained category diversity, and narrow scene coverage. To address these limitations, we introduce XS-VID, a large-scale video benchmark comprising 223K frames and 1.4M annotated bounding boxes across 374 video sequences spanning diverse scene types. XS-VID provides extensive coverage of small-object scales, particularly for extremely small ($0\sim12^2$ pixels) and small ($12^2\sim20^2$ pixels) objects, which collectively constitute over 55% of all annotations. For systematic evaluation, we establish three dedicated tracks: Detection, multiple object tracking (MOT), and single object tracking (SOT), and extensively test the existing state-of-the-art methods on each. The experimental results indicate that existing methods face significant challenges with XS-VID, mainly stemming from insufficient modeling of spatiotemporal features at small scales. To tackle these challenges, we propose a lightweight, high-precision detection framework dubbed YOLOFT. It enhances small-object feature representation and spatiotemporal integration while preserving high detection speed, thereby achieving improved accuracy and robustness on both the XS-VID and VisDrone benchmarks. Our dataset and code are publicly available at https://gjhhust.github.io/XS-VID/, providing a solid foundation for future research on small-object detection and tracking in videos.

cs.CV↗

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

Vector Quantization (VQ) is fundamental to discrete visual tokenizers that power modern autoregressive and masked image generation models. While recent shared-projection codebook methods have substantially advanced codebook utilization, training stability remains a critical and underexplored challenge. We argue that the root cause lies in the entanglement of the Encoder--Decoder and Codebook training: because neither module can reliably fulfill its own responsibility in isolation, the system can only function when the two subsystems happen to cooperate---a fragile condition that breaks down precisely when training is most stressed. We propose StableVQ, which revisits the proper learning objective of each module and resolves the problems that arise when each is trained to fulfill its own role independently. Concretely, (1) Dynamic STE corrects the instability in the Encoder's learning objective, enabling it to robustly optimize the reconstruction space under discrete regularization even when codebook utilization is low. (2) Region VQ Loss reconceives the Codebook's learning objective so that it can independently guarantee full tracking of the encoder output distribution, without relying on encoder oscillations to drive activation. (3) Decoupled Schedule recognizes that the distinct responsibilities of the Encoder--Decoder and the Codebook demand distinct optimization dynamics, and assigns each an independent learning rate schedule to ensure robust system-level behavior. Built on top of shared-projection codebooks, StableVQ is lightweight and introduces no learnable parameters. Experiments on ImageNet demonstrate consistent improvements in training stability, codebook utilization, and reconstruction quality across diverse codebook sizes and initialization settings.

cs.CV↗

Chiral Packings in Cylinders are Ultrasensitive to Confinement Deformation

Sphere packings in circular cylinders have attracted substantial research interest, among which the discovery of chiral helical structures is the most iconic. However, recent experimental results on zebrafish do not match the known packing structures in circular cylinders. To account for the inherent imperfections of biological tubes, we take elliptic cylinders as the canonical deformation of circular cylinders and investigate the densest packings of hard spheres in them using simulation, theory, and experiments. Starting from the chiral structures in circular cylinders, we demonstrate that even a weak cross-sectional deformation can trigger entirely new phases, including ones that either eliminate global chirality or significantly complicate the chiral structures. This reveals the significant effect of cylindrical anisotropy. The new helical phases under anisotropic confinement remain chiral and develop hierarchical periodic structures, which are difficult to obtain by simulations but are predicted by our newly developed theory for helical phases in elliptic cylinders. The theory also predicts double oscillated-chain phases without chirality, which perfectly match the simulations. Our work offers fresh insights into understanding packings in anisotropic cylinders, which will help researchers to design new materials and to understand many living systems.

cond-mat.soft↗

VQRAE: Representation Quantization Autoencoders for Multimodal Understanding, Generation and Reconstruction

Unifying multimodal understanding, generation and reconstruction representation in a single tokenizer remains a key challenge in building unified models. Previous research predominantly attempts to address this in a dual encoder paradigm, e.g., utilizing the separate encoders for understanding and generation respectively or balancing semantic representations and low-level features with contrastive loss. In this paper, we propose VQRAE, a Vector Quantization version of Representation AutoEncoders, which pioneers the first exploration in unified representation to produce Continuous semantic features for image understanding and Discrete tokens for visual generation within a unified tokenizer. Specifically, we build upon pretrained vision foundation models with a symmetric ViT decoder and adopt a two-stage training strategy: first, it freezes the encoder and learns a high-dimensional semantic VQ codebook with pixel reconstruction objective; then jointly optimizes the encoder with self-distillation constraints. This design enables negligible semantic information for maintaining the ability of multimodal understanding, discrete tokens that are compatible for generation and fine-grained reconstruction. Besides, we identify the intriguing property in quantizing semantic encoders that rely on high-dimensional codebook in contrast to the previous common practice of low-dimensional codebook in image reconstruction. The semantic VQ codebook can achieve a 100% utilization ratio at a dimension of 1536. VQRAE presents competitive performance on several benchmarks of visual understanding, generation and reconstruction with promising scaling property in the autoregressive paradigm for its discrete merits.

cs.CV↗

Visual Generation Tuning

Large Vision Language Models (VLMs) effectively bridge the modality gap through extensive pretraining, acquiring sophisticated visual representations aligned with language. However, it remains underexplored whether these representations, optimized for multimodal understanding tasks, harbor an inherent potential for visual generation. In this paper, we propose VGT, Visual Generation Tuning, a novel paradigm designed to stimulate the underlying capabilities of visual generation within any vision language models. By performing efficient visual generation tuning on well-pretrained VLMs, we significantly mitigate the alignment costs and accelerate the convergence of autoregressive modeling in the continuous space (20x speedup). Specifically, we dismiss the entangled pixel-level VAEs designed for diffusion transformers and formulate VGT-AE through aligning the semantic encoders from pretrained VLMs with the latent representations of pixel decoders. In image reconstruction tasks, we achieve 26.67 PSNR and 0.50 rFID at a 28x compression ratio, outperforming specialized VAEs; in visual generation tasks, we achieve state-of-the-art outcomes among autoregressive models, 0.77 on GenEval and 78.73 on DPG-Bench. Furthermore, our proposed VGT showcases significant scaling promise and is versatile for endowing any VLMs trained for multimodal understanding with the capabilities of visual generation, which paves the new avenue to explore next-generation unified multimodal foundation models. Models and codes are available at https://github.com/hustvl/VGT.

cs.CV↗

CRouting: Reducing Expensive Distance Calls in Graph-Based Approximate Nearest Neighbor Search

Approximate nearest neighbor search (ANNS) is a crucial problem in information retrieval and AI applications. Recently, there has been a surge of interest in graph-based ANNS algorithms due to their superior efficiency and accuracy. However, the repeated computation of distances in high-dimensional spaces constitutes the primary time cost of graph-based methods. To accelerate the search, we propose a novel routing strategy named CRouting, which bypasses unnecessary distance computations by exploiting the angle distributions of high-dimensional vectors. CRouting is designed as a plugin to optimize existing graph-based search with minimal code modifications. Our experiments show that CRouting reduces the number of distance computations by up to 41.5% and boosts queries per second by up to 1.48$\times$ on two predominant graph indexes, HNSW and NSG. Code is publicly available at https://github.com/ISCS-ZJU/CRouting.

cs.DB↗

High-Performance Privacy-Preserving Matrix Completion for Trajectory Recovery

Matrix completion has important applications in trajectory recovery and mobile social networks. However, sending raw data containing personal, sensitive information to cloud computing nodes may lead to privacy exposure issue.The privacy-preserving matrix completion is a useful approach to perform matrix completion while preserving privacy. In this paper, we propose a high-performance method for privacy-preserving matrix completion. First,we use a lightweight encryption scheme to encrypt the raw data and then perform matrix completion using alternating direction method of multipliers (ADMM). Then,the complemented matrix is decrypted and compared with the original matrix to calculate the error. This method has faster speed with higher accuracy. The results of numerical experiments reveal that the proposed method is faster than other algorithms.

cs.CR↗

Circuit as Set of Points

As the size of circuit designs continues to grow rapidly, artificial intelligence technologies are being extensively used in Electronic Design Automation (EDA) to assist with circuit design. Placement and routing are the most time-consuming parts of the physical design process, and how to quickly evaluate the placement has become a hot research topic. Prior works either transformed circuit designs into images using hand-crafted methods and then used Convolutional Neural Networks (CNN) to extract features, which are limited by the quality of the hand-crafted methods and could not achieve end-to-end training, or treated the circuit design as a graph structure and used Graph Neural Networks (GNN) to extract features, which require time-consuming preprocessing. In our work, we propose a novel perspective for circuit design by treating circuit components as point clouds and using Transformer-based point cloud perception methods to extract features from the circuit. This approach enables direct feature extraction from raw data without any preprocessing, allows for end-to-end training, and results in high performance. Experimental results show that our method achieves state-of-the-art performance in congestion prediction tasks on both the CircuitNet and ISPD2015 datasets, as well as in design rule check (DRC) violation prediction tasks on the CircuitNet dataset. Our method establishes a bridge between the relatively mature point cloud perception methods and the fast-developing EDA algorithms, enabling us to leverage more collective intelligence to solve this task. To facilitate the research of open EDA design, source codes and pre-trained models are released at https://github.com/hustvl/circuitformer.

cs.CV↗

Resolution limit revisited: community detection using generalized modularity density

Various attempts have been made in recent years to solve the Resolution Limit (RL) problem in community detection by considering variants of the modularity metric in the detection algorithms. These metrics purportedly largely mitigate the RL problem and are preferable to modularity in many realistic scenarios. However, they are not generally suitable for analyzing weighted networks or for detecting hierarchical community structure. Resolution limit problems can be complicated, though, and in particular it can be unclear when it should be considered as problem. In this paper, we introduce a metric that we call generalized modularity density $Q_g$ that eliminates the RL problem at any desired resolution and is easily extendable to study weighted and hierarchical networks. We also propose a benchmark test to quantify the resolution limit problem, examine various modularity-like metrics to show that the new metric $Q_g$ performs best, and show that $Q_g$ can identify modular structure in real-world and artificial networks that is otherwise hidden.

physics.soc-ph↗

Reduced network extremal ensemble learning (RenEEL) scheme for community detection in complex networks

We introduce an ensemble learning scheme for community detection in complex networks. The scheme uses a Machine Learning algorithmic paradigm we call Extremal Ensemble Learning. It uses iterative extremal updating of an ensemble of network partitions, which can be found by a conventional base algorithm, to find a node partition that maximizes modularity. At each iteration, core groups of nodes that are in the same community in every ensemble partition are identified and used to form a reduced network. Partitions of the reduced network are then found and used to update the ensemble. The smaller size of the reduced network makes the scheme efficient. We use the scheme to analyze the community structure in a set of commonly studied benchmark networks and find that it outperforms all other known methods for finding the partition with maximum modularity.

physics.soc-ph↗

A Note On the Rank of the Optimal Matrix in Symmetric Toeplitz Matrix Completion Problem

We consider the symmetric Toeplitz matrix completion problem, whose matrix under consideration possesses specific row and column structures. This problem, which has wide application in diverse areas, is well-known to be computationally NP-hard. This note provides an upper bound on the objective of minimizing the rank of the symmetric Toeplitz matrix in the completion problem based on the theorems from the trigonometric moment problem and semi-infinite problem. We prove that this upper bound is less than twice the number of linear constraints of the Toeplitz matrix completion problem. Compared with previous work in the literature, ours is one of the first efforts to investigate the bound of the objective value of the Toeplitz matrix completion problem.

math.OC↗