Searcharxiv⌕ Search

arXiv subjects

Xinyi Guo

Publications and source records attributed to Xinyi Guo.

At least 19 recordsLinked to original sources

MI-Hyper: Generating Multiple Instance Models for Weight Space Transfer Learning in Histopathology via Mixed Tasks Pretrained HyperNetwork

Zero-shot whole-slide image (WSI) prediction aims to generalize to new pathology tasks without parameter updates at deployment. MI-Zero makes these predictions by pooling patch-text similarities. Conditional weight generation, combined with supervision from diverse pathology tasks, offers an alternative: learning to construct predictive rules from task descriptions. We introduce MI-Hyper, a hypernetwork pretrained across mixed tasks to generate the aggregation and classification weights of multiple instance learning (MIL) models. We first train the generator and a shared MIL basis jointly using source labels. We then freeze the basis and refine the generator. The generated task weights are reused across cases. Across 308 tasks from 52 pathology datasets, MI-Hyper improves mean AUC, balanced accuracy and macro-F1 over MI-Zero by 1.85, 1.17 and 0.57 percentage points, respectively. On some molecular prediction tasks, both AUC and balanced accuracy improve by more than 20 points. These results support task-conditioned weight generation as a promising approach to weight-space transfer for long-sequence modeling.

cs.CV↗

Unlocking Fine-Grained Cell Representations via Soft-Interior Rim Dense-Grid Pooling in Pathology Foundation Models

Fine-grained cell representations are essential for characterizing cellular heterogeneity and tissue organization in computational pathology. Pathology foundation models (PFMs) provide powerful pretrained features for transferable tissue analysis. HoVer-Net uses task-trained segmentation and typing, whereas CellViT++ averages native-grid tokens assigned to detected nuclei. However, the former requires cell-specific training; the latter ignores fractional nuclear coverage and can mix nuclear content with neighbouring tissue. Therefore, we propose SRDP, short for Soft-interior Rim Dense-grid Pooling, a training-free geometry-aware readout. We apply the frozen patch projection at half stride, interleave four offset lattices, map the cell mask to fractional coverage weights, and concatenate whole-cell pooling with two inner boundary pools. Across four datasets and six frozen encoders, SRDP improves mean balanced accuracy by 5.68 percentage points over native token averaging across 24 settings, including 11.22 points on average for CoNIC and Lizard. SRDP converts a coarse frozen PFM lattice into a mask-aligned, fine-grained cell representation without updating the encoder.

cs.CV↗

PRELUDE: Separating Common Components from Target-Specific Dependencies for Multivariate Time Series Forecasting

Multivariate time series forecasting often improves the prediction of a target by modeling the interactions among variables, and cross-variable modules such as cross-sectional pooling, inter-variable attention, and graph structures are used to construct such interactions. However, by intervening on the cross-variable reading of existing models at inference time, we find that although these models substantially reduce forecasting error through cross-variable interaction, their reading is nearly identical for all targets: the gain appears to come from dependencies among variables but mainly comes from the common component. We term this phenomenon spurious dependency and attribute it to common-component confounding: present in every variable, the common component affects how reading distinguishes sources, so that reading tends to be shared across targets and concentrates on a few similar averages; dependencies thus make up a small share of the read-in content, and a common component composed of several drivers is only partly captured. Theoretical analysis further shows that the common component and dependencies place different requirements on reading and should therefore be handled in different ways. Accordingly, we propose PRELUDE, a separated-reading framework that estimates the common component holistically through cross-sectional pooling and multiple common factors, and reads dependencies for each target via lead-lag relations in the innovation space obtained by prewhitening. Extensive experiments on 16 real-world datasets from four domains show that PRELUDE effectively mitigates spurious dependency and achieves the best overall performance: it attains the best MSE on 23 of the 24 short-horizon and air-quality settings and reduces the average MSE by 6.1% relative to the strongest baseline.

cs.LG↗

FreshCast: Timely and Complementary Retrieval for Frozen Time Series Forecasters

Retrieval-augmented forecasting for Web and online services supplements a forecaster with the continuations of historical segments that resemble the current context. Most existing methods build the retrieval memory once from the training segment, leaving observations revealed after deployment unavailable as references, and generally do not calibrate how much retrieved information should influence a frozen forecaster. We identify two key determinants of retrieval utility for a frozen forecaster: the timeliness of the retrieved history and its complementarity, i.e., whether the correction it induces aligns with the forecaster's residual errors; this alignment can shift between validation and deployment when the memory becomes stale. We propose FreshCast, a plug-in retrieval framework for frozen forecasters: a non-parametric streaming memory keeps writing in new observations, relational kernel regression weights candidate readouts into a memory forecast, and its correction weight is calibrated in closed form on the validation segment. Under a simplified generative model, we characterize the optimal combination gain through the second-order relation between forecaster error and memory correction, and show that periodic memory information can become redundant once the forecaster captures it. Across seven datasets, ten forecasters, and three input lengths, FreshCast reduces average MSE for every forecaster and input length, by 14.0% and 5.7% at L = 96 and 720, and achieves lower average MSE than the evaluated retrieval-augmented and online updating baselines in their comparison settings. Ablations show that post-training observations are a primary source of the gain: freezing the memory at the end of training removes most of it. For a frozen forecaster, similarity alone is not enough: useful retrieval must stay timely and provide information that the forecaster still misses.

cs.LG↗

The Car Resequencing Problem with Makespan Minimization

Problem definition: The car resequencing problem involves rearranging the sequence of cars between two adjacent production shops via an intermediate buffer. This study explores a new variant that aims to minimize the resequencing makespan when the buffer features several forward lanes and a return lane (RL). Both types of lanes follow first-in-first-out rules, but operate in opposite directions. Through the RL, cars can circulate between forward lanes, thereby increasing resequencing flexibility and complexity. The goal is to position cars over time to complete the target sequence change in the shortest possible time. Methodology/results: We prove the problem is NP-complete and develop an exact two-stage branch-and-price approach integrated with constraint programming (BP-CP). Three versions of the master problem (MP) are proposed for the first stage, each paired with a subproblem defined on a specific graph. Using partial integer solutions from branching, a constraint-programming model built in the second stage yields either a feasible solution to the problem or a feasibility cut added to the MP. Experimental results indicate that BP-CP greatly increases the scale of solvable instances for a production batch, especially with a moderately relaxed MP. A comparative analysis shows that the RL markedly enhances resequencing flexibility, enabling smaller buffers to attain makespans comparable to those of larger buffers without the RL across varying resequencing complexities. Managerial implications: Expanding lane capacity alone yields limited gains in resequencing performance and may even reduce efficiency. The RL acts as several forward lanes, supporting a smaller and cost-effective buffer design while maintaining efficient just-in-time operations.

math.OC↗

SlideCheck: Guiding Self-Supervised Pretraining of Pathology Foundation Models via Dataset Distributions

Pathology foundation models are pretrained on large streams of WSI-derived patches, while supervision during data construction is often slide-level, sparse, or heterogeneous. This mismatch makes it difficult to understand and control which biological patterns enter the pretraining data. We propose SlideCheck, a lightweight pretraining data guidance tool built on frozen pathology foundation model patch features. Rather than serving as a standalone patch diagnostic model, SlideCheck provides explicit abnormality and malignancy scores for organizing, filtering, and auditing pathology pretraining data. SlideCheck uses a dual-head MLP to separately model broad abnormal morphology and malignant evidence. A regularized feature-space scorer provides a supervised anchor for patch-level evidence estimation, while score-attention agreement combines patch scores with WSI-level MIL attention to mine high-confidence pseudo labels. The same scores are then used to construct broad-positive ViT pretraining subsets, where a patch is selected if either abnormality or malignancy evidence exceeds a threshold. Experiments show that SlideCheck-defined data distributions influence the downstream behavior of self-supervised ViT pretraining, indicating that biological composition is an important controllable factor in pathology foundation model development. Curated subsets can approach full-data performance, suggesting that explicitly scored patch pools may support more efficient and auditable pretraining data construction. These findings position SlideCheck as a data guidance and auditing layer for transforming large, undifferentiated patch pools into controllable and reusable pretraining datasets.

cs.CV↗

Is Class Signal Clustered or Routed in Task-Induced Implicit Neural Representation Weight Spaces?

Implicit neural representations (INRs) encode images as neural-network weights, making image classification a problem of weight-space classifiability. A natural geometric hypothesis is that classifier feedback should make image-specific weights cluster by class in the shared-anchor coordinate. We test this hypothesis in the SIREN-based Meta Weight Transformer (MWT) regime, where end-to-end training meta-learns a shared initialization and inner-loop update schedule for fitting image-specific SIRENs. We find that this prediction fails. Exposed weight-space geometry and supervised clustering pressure do not reliably track trained-reader accuracy; clustering can even make local neighborhoods more class-consistent while making the trained reader worse. Crucially, the reader constructs rather than inherits class-aligned geometry: token-flow diagnostics show that class-aligned neighborhoods become strongly predictive of trained-reader accuracy only after late reader interactions, not in the input coordinate. We further identify the native SIREN bias column in the augmented weight token as a low-dimensional, sample-dependent causal readout route for the trained reader; targeted controls rule out generic scalar-column and marginal-distribution artifacts. The diagnosis motivates interventions that strengthen reader routing, add an explicit bias route, or use denser inner-loop fitting; under the lane-specific training conventions used here, route-directed variants often outperform the shared-anchor baseline but interact non-additively. Task-induced INR weights are classifiable not because they form raw geometric clusters, but because their class signal is routed through the reader.

cs.CV↗

Uplink RSMA Performance Analysis with Rate Adaptation: A Stochastic Geometry Approach

Rate-splitting multiple access (RSMA) has emerged as a promising technique for efficient interference management in next-generation wireless networks. While most existing studies focus on downlink and single-cell designs, the modeling and analysis of uplink RSMA under large-scale deployments remain largely unexplored. On the basis of stochastic geometry (SG), this paper introduces a unified analytical framework that integrates finite modulation and coding scheme (MCS)-based rate adaptation. This framework jointly captures spatial interference coupling and discrete rate behavior to bridge theoretical tractability and practical realism. Within this framework, we derive tractable expressions for the conditional received rate (CRR), its spatial average, and higher-order statistics via the meta distribution, thereby quantifying both the mean and user-specific rate performance. Results show that the proposed unified framework not only generalizes existing non-orthogonal multiple access (NOMA) and orthogonal multiple access (OMA) analyses but also provides new insights into how discrete rate adaptation reshapes interference dynamics and fairness in dense RSMA-enabled networks.

cs.IT↗

SOME: Symmetric One-Hot Matching Elector -- A Lightweight Microsecond Decoder for Quantum Error Correction

Conventional quantum error correction (QEC) decoders such as Minimum-Weight Perfect Matching (MWPM) and Union-Find (UF) offer high thresholds and fast decoding, respectively, but both suffer from high topological complexity. In contrast, Ising model-based decoders reduce topological complexity but demand considerable decoding time. We propose the Symmetric One-Hot Matching Elector (SOME), a novel decoder that reformulates the QEC decoding task as a Quadratic Unconstrained Binary Optimization (QUBO) problem -- termed the One-Hot QUBO (OHQ). Each variable in the QUBO represents whether a given pair of flipped syndromes is matched, while the error probabilities between the pair are encoded as interaction coefficients (weight). Constraints ensure that each flipped syndrome is matched exactly once. Valid solutions of OHQ correspond to self-inverse permutation matrices, characterized by symmetric one-hot encoding. To solve the OHQ efficiently, SOME reformulates the decoding task as the construction of permutation matrices that minimize the total weight. It initializes each candidate matrix from one of the minimum-weight syndrome pairs, then iteratively appends additional pairs in ascending order of weight, and finally selects the permutation matrix with the lowest total energy. SOME achieves up to a 99.9x reduction in variable count and reduces decoding times from milliseconds to microseconds on a single-threaded commodity CPU. OHQ also maintains performance up to a 10.5% physical error rate, surpassing the highest known threshold of MWPM@.

cs.ET↗

MDIT-Bench: Evaluating the Dual-Implicit Toxicity in Large Multimodal Models

The widespread use of Large Multimodal Models (LMMs) has raised concerns about model toxicity. However, current research mainly focuses on explicit toxicity, with less attention to some more implicit toxicity regarding prejudice and discrimination. To address this limitation, we introduce a subtler type of toxicity named dual-implicit toxicity and a novel toxicity benchmark termed MDIT-Bench: Multimodal Dual-Implicit Toxicity Benchmark. Specifically, we first create the MDIT-Dataset with dual-implicit toxicity using the proposed Multi-stage Human-in-loop In-context Generation method. Based on this dataset, we construct the MDIT-Bench, a benchmark for evaluating the sensitivity of models to dual-implicit toxicity, with 317,638 questions covering 12 categories, 23 subcategories, and 780 topics. MDIT-Bench includes three difficulty levels, and we propose a metric to measure the toxicity gap exhibited by the model across them. In the experiment, we conducted MDIT-Bench on 13 prominent LMMs, and the results show that these LMMs cannot handle dual-implicit toxicity effectively. The model's performance drops significantly in hard level, revealing that these LMMs still contain a significant amount of hidden but activatable toxicity. Data are available at https://github.com/nuo1nuo/MDIT-Bench.

cs.CL↗

"I Feel Myself So Small!": Designing and Evaluating VR Awe Experiences Based on Theories Related to Sublime

Research suggests the potential of employing VR to elicit awe experiences, thereby promoting well-being. Building upon theories related to the sublime and embodiment, we designed three VR scenes to evaluate the effectiveness of sublime and embodied design elements in invoking awe experiences. We conducted a within-subject study involving 28 young adults who experienced the three VR designs. Results demonstrated that the VR design with sublime elements significantly elicited more intense awe experiences compared to the one without, while adding embodied elements did not enhance the intensity of awe. Qualitative interviews revealed critical design elements (e.g., the obscure event should be reasonable) and their underlying mechanisms (e.g., leading to feelings of enlightenment) in invoking awe experiences. We further discuss considerations and implications for the design of effective awe-inspiring VR applications.

cs.HC↗

Operation Scheme Optimizations to Achieve Ultra-high Endurance (1010) in Flash Memory with Robust Reliabilities

Flash memory has been widely adopted as stand-alone memory and embedded memory due to its robust reliability. However, the limited endurance obstacles its further applications in storage class memory (SCM) and to proceed endurance-required computing-in-memory (CIM) tasks. In this work, the optimization strategies have been studied to tackle this concern. It is shown that by adopting the channel hot electrons injection (CHEI) and hot hole injection (HHI) to implement program/erase (PE) cycling together with a balanced memory window (MW) at the high-Vth (HV) mode, impressively, the endurance can be greatly extended to 1010 PE cycles, which is a record-high value in flash memory. Moreover, by using the proposed electric-field-assisted relaxation (EAR) scheme, the degradation of flash cells can be well suppressed with better subthreshold swings (SS) and lower leakage currents (sub-10pA after 1010 PE cycles). Our results shed light on the optimization strategy of flash memory to serve as SCM and implementendurance-required CIM tasks.

eess.SY↗

Efficient Fourier single-pixel imaging with Gaussian random sampling

Fourier single-pixel imaging (FSI) is a branch of single-pixel imaging techniques. It uses Fourier basis patterns as structured patterns for spatial information acquisition in the Fourier domain. However, the spatial resolution of the image reconstructed by FSI mainly depends on the number of Fourier coefficients sampled. The reconstruction of a high-resolution image typically requires a number of Fourier coefficients to be sampled, and therefore takes a long data acquisition time. Here we propose a new sampling strategy for FSI. It allows FSI to reconstruct a clear and sharp image with a reduced number of measurements. The core of the proposed sampling strategy is to perform a variable density sampling in the Fourier space and, more importantly, the density with respect to the importance of Fourier coefficients is subject to a one-dimensional Gaussian function. Combined with compressive sensing, the proposed sampling strategy enables better reconstruction quality than conventional sampling strategies, especially when the sampling ratio is low. We experimentally demonstrate compressive FSI combined with the proposed sampling strategy is able to reconstruct a sharp and clear image of 256-by-256 pixels with a sampling ratio of 10%. The proposed method enables fast single-pixel imaging and provides a new approach for efficient spatial information acquisition.

eess.IV↗

COVID-19 Contact-tracing Apps: a Survey on the Global Deployment and Challenges

To address the massive spike in uncertainties triggered by the coronavirus disease (COVID-19), there is an ever-increasing number of national governments that are rolling out contact-tracing Apps to aid the containment of the virus. The first hugely contentious issue facing the Apps is the deployment framework, i.e. centralized or decentralized. Based on this, the debate branches out to the corresponding technologies that underpin these architectures, i.e. GPS, QR codes, and Bluetooth. This work conducts a pioneering review of the above scenarios and contributes a geolocation mapping of the current deployment. The Apps vulnerabilities and the directions of research are identified, with a special focus on the Bluetooth-inspired decentralized paradigm.

cs.CR↗

A Novel Twitter Sentiment Analysis Model with Baseline Correlation for Financial Market Prediction with Improved Efficiency

A novel social networks sentiment analysis model is proposed based on Twitter sentiment score (TSS) for real-time prediction of the future stock market price FTSE 100, as compared with conventional econometric models of investor sentiment based on closed-end fund discount (CEFD). The proposed TSS model features a new baseline correlation approach, which not only exhibits a decent prediction accuracy, but also reduces the computation burden and enables a fast decision making without the knowledge of historical data. Polynomial regression, classification modelling and lexicon-based sentiment analysis are performed using R. The obtained TSS predicts the future stock market trend in advance by 15 time samples (30 working hours) with an accuracy of 67.22% using the proposed baseline criterion without referring to historical TSS or market data. Specifically, TSS's prediction performance of an upward market is found far better than that of a downward market. Under the logistic regression and linear discriminant analysis, the accuracy of TSS in predicting the upward trend of the future market achieves 97.87%.

cs.SI↗

Electron Heating in Low Mach Number Perpendicular shocks. II. Dependence on the Pre-Shock Conditions

Recent X-ray observations of merger shocks in galaxy clusters have shown that the post-shock plasma is two-temperature, with the protons being hotter than the electrons. In this work, the second of a series, we investigate by means of two-dimensional particle-in-cell simulations the efficiency of electron irreversible heating in perpendicular low Mach number shocks. We consider values of plasma beta (ratio of thermal and magnetic pressures) in the range $4\lesssim β_{p0}\lesssim 32$ and sonic Mach number (ratio of shock speed to pre-shock sound speed) in the range $2\lesssim M_{s}\lesssim 5$, as appropriate for galaxy cluster shocks. As shown in Paper I, magnetic field amplification - induced by shock compression of the pre-shock field, or by strong proton cyclotron and mirror modes accompanying the relaxation of proton temperature anisotropy - can drive the electron temperature anisotropy beyond the threshold of the electron whistler instability. The growth of whistler waves breaks the electron adiabatic invariance, and allows for efficient entropy production. We find that the post-shock electron temperature $T_{e2}$ exceeds the adiabatic expectation $T_{e2,\rm ad}$ by an amount $(T_{e2}-T_{e2,\rm ad})/T_{e0}\simeq 0.044 \,M_s (M_s-1)$ (here, $T_{e0}$ is the pre-shock temperature), which depends only weakly on the plasma beta, over the range $4\lesssim β_{p0}\lesssim 32$ which we have explored, and on the proton-to-electron mass ratio (the coefficient of $\simeq 0.044$ is measured for our fiducial $m_i/m_e=49$, and we estimate that it will decrease to $\simeq 0.03$ for the realistic mass ratio). Our results have important implications for current and future observations of galaxy cluster shocks in the radio band (synchrotron emission and Sunyaev-Zel'dovich effect) and at X-ray frequencies.

astro-ph.HE↗

Electron Heating in Low Mach Number Perpendicular shocks. I. Heating Mechanism

Recent X-ray observations of merger shocks in galaxy clusters have shown that the post-shock plasma is two-temperature, with the protons hotter than the electrons. By means of two-dimensional particle-in-cell simulations, we study the physics of electron irreversible heating in perpendicular low Mach number shocks, for a representative case with sonic Mach number of 3 and plasma beta of 16. We find that two basic ingredients are needed for electron entropy production: (i) an electron temperature anisotropy, induced by field amplification coupled to adiabatic invariance; and (ii) a mechanism to break the electron adiabatic invariance itself. In shocks, field amplification occurs at two major sites: at the shock ramp, where density compression leads to an increase of the frozen-in field; and farther downstream, where the shock-driven proton temperature anisotropy generates strong proton cyclotron and mirror modes. The electron temperature anisotropy induced by field amplification exceeds the threshold of the electron whistler instability. The growth of whistler waves breaks the electron adiabatic invariance, and allows for efficient entropy production. We find that the electron heating efficiency displays only a weak dependence on mass ratio (less than 30 percent drop, as we increase the mass ratio from 49 up to 1600). We develop an analytical model of electron irreversible heating and show that it is in excellent agreement with our simulation results.

astro-ph.HE↗

Periodic Signals in Binary Microlensing Events

Gravitational microlensing events are powerful tools for the study of stellar populations. In particular, they can be used to discover and study a variety of binary systems. A large number of binary lenses have already been found through microlensing surveys and a few of these systems show strong evidence of orbital motion on the timescale of the lensing event. We expect that more binary lenses of this kind will be detected in the future. For binaries whose orbital period is comparable to the event duration, the orbital motion can cause the lensing signal to deviate drastically from that of a static binary lens. The most striking property of such light curves is the presence of quasi- periodic features, which are produced as the source traverses the same regions in the rotating lens plane. These repeating features contain information about the orbital period of the lens. If this period can be extracted, then much can be learned about the lensing system even without performing time-consuming, detailed light curve modeling. However, the relative transverse motion between the source and the lens significantly complicates the problem of period extraction. To resolve this difficulty, we present a modification of the standard Lomb-Scargle periodogram analysis. We test our method for four representative binary lens systems and demonstrate its efficiency in correctly extracting binary orbital periods.

astro-ph.GA↗