Searcharxiv⌕ Search

arXiv subjects

Yi Zhang

Publications and source records attributed to Yi Zhang.

At least 271 records · Page 15Linked to original sources
arXiv2026

P$^2$HCT: Plug-and-Play Hierarchical C2F Transformer for Multi-Scale Feature Fusion

Feature fusion plays a pivotal role in achieving high performance in vision models, yet existing attention-based fusion techniques often suffer from substantial computational overhead and implementation complexity, particularly in resource-constrained settings. To address these limitations, we introduce the Plug-and-Play Hierarchical C2F Transformer (P$^2$HCT), a lightweight module that combines coarse-to-fine token selection with shared attention parameters to preserve spatial details while reducing inference cost. P$^2$HCT is trainable using coarse attention alone and can be seamlessly activated at inference to enhance accuracy without retraining. Integrated into real-time detectors such as YOLOv11-N/S/M, P$^2$HCT achieves mAP gains of 0.9\%, 0.5\%, and 0.4\% on MS COCO with minimal latency increase. Similarly, embedding P$^2$HCT into ResNet-18/50/101 backbones improves ImageNet top-1 accuracy by 6.5\%, 1.7\%, and 1.0\%, respectively. These results underscore P$^2$HCT's effectiveness as a hardware-friendly and general-purpose enhancement for both detection and classification tasks.

cs.CV↗
arXiv2026

Unbiased Multimodal Reranking for Long-Tail Short-Video Search

Kuaishou serving hundreds of millions of searches daily, the quality of short-video search is paramount. However, it suffers from a severe Matthew effect on long-tail queries: sparse user behavior data causes models to amplify low-quality content such as clickbait and shallow content. The recent advancements in Large Language Models (LLMs) offer a new paradigm, as their inherent world knowledge provides a powerful mechanism to assess content quality, agnostic to sparse user interactions. To this end, we propose a LLM-driven multimodal reranking framework, which estimates user experience without real user behavior. The approach involves a two-stage training process: the first stage uses multimodal evidence to construct high-quality annotations for supervised fine-tuning, while the second stage incorporates pairwise preference optimization to help the model learn partial orderings among candidates. At inference time, the resulting experience scores are used to promote high-quality but underexposed videos in reranking, and further guide page-level optimization through reinforcement learning. Experiments show that the proposed method achieves consistent improvements over strong baselines in offline metrics including AUC, NDCG@K, and human preference judgement. An online A/B test covering 15\% of traffic further demonstrates gains in both user experience and consumption metrics, confirming the practical value of the approach in long-tail video search scenarios.

cs.IR↗
arXiv2026

The average X-ray spectrum of the volume-complete M-, F-, G-, and K-type star sample within 10 pc of the Sun

F, G, K and M type stars are the most abundant stellar population in the Milky Way and are expected to contribute to its diffuse X-ray emission. Yet their intrinsic average X-ray spectrum remains poorly constrained due to their faint X-ray luminosities, leaving their collective role in the X-ray background of the Milky Way uncertain. We analysed the volume-complete sample of M- (M0--M6) and FGK-type stars within 10 pc of the Sun using data from eROSITA all-sky survey aboard the Spectrum-Roentgen-Gamma (SRG) mission (eRASS:4). Individual spectra were normalized by exposure and distance and stacked to produce representative averages. The distance-normalized emission measures yield an average X-ray luminosity of $(2.6 \pm0.1)\times 10^{27}$ erg/s for M-type stars, and $(15\pm3)\times 10^{27}$ erg/s for F, G and K-type stars in 0.2--2.0 keV. The average spectra could be well described by a sum of three and two thermal models. Fitted temperatures and abundances remain consistent across M-star subgroups, while early-M stars are surprisingly on average less luminous than mid/late-M types. These results offer new insights into the collective X-ray properties of nearby stars, and provide motivation to explore the link with the unresolved soft X-ray background of the Galaxy.

astro-ph.HE↗
arXiv2026

AffordGrasp: Cross-Modal Diffusion for Affordance-Aware Grasp Synthesis

Generating human grasping poses that accurately reflect both object geometry and user-specified interaction semantics is essential for natural hand-object interactions in AR/VR and embodied AI. However, existing semantic grasping approaches struggle with the large modality gap between 3D object representations and textual instructions, and often lack explicit spatial or semantic constraints, leading to physically invalid or semantically inconsistent grasps. In this work, we present AffordGrasp, a diffusion-based framework that produces physically stable and semantically faithful human grasps with high precision. We first introduce a scalable annotation pipeline that automatically enriches hand-object interaction datasets with fine-grained structured language labels capturing interaction intent. Building upon these annotations, AffordGrasp integrates an affordance-aware latent representation of hand poses with a dual-conditioning diffusion process, enabling the model to jointly reason over object geometry, spatial affordances, and instruction semantics. A distribution adjustment module further enforces physical contact consistency and semantic alignment. We evaluate AffordGrasp across four instruction-augmented benchmarks derived from HO-3D, OakInk, GRAB, and AffordPose, and observe substantial improvements over state-of-the-art methods in grasp quality, semantic accuracy, and diversity.

cs.RO↗
arXiv2026

Many-body Josephson diode effect in superconducting quantum interferometers

We propose a many-body mechanism for a strong Josephson diode effect (JDE) in an interacting nanoscale SQUID formed by two parallel quantum dots coupled to superconducting leads. Unlike conventional diode behavior, where nonreciprocity originates from a skewed current-phase relation within a single, continuously evolving ground state, the JDE reported here is \emph{branch selected}: the positive and negative critical currents are optimized on different many-body branches across the $0$-$π$ phase boundary, yielding a substantial enhancement of the diode efficiency. We further show that a \emph{nonlocal} Cooper-pair tunneling channel, which binds the two electrons on different arms, is essential: it reshapes the $0$-$π$ boundary and produces a pronounced ``diode band'' in parameter space, in sharp contrast to the fragile hotspot obtained when only local Cooper-pair transfer is available. While the key physics is captured by an effective model in the superconducting atomic limit, our conclusions remain robust for realistic finite-gap devices, as demonstrated within a generalized atomic-limit framework.

cond-mat.supr-con↗
arXiv2026

BrainRing: An Interactive Web-Based Tool for Brain Connectivity Chord Diagram Visualization

Visualizing brain functional connectivity (FC) patterns is essential for understanding neural organization, yet existing tools such as Circos and BrainNet Viewer require complex configuration files or proprietary software environments. We present BrainRing, a free, open-source, browser-based interactive tool for generating publication-quality chord diagrams of brain connectivity data. BrainRing requires no installation, backend server, or programming knowledge. Users simply open a single HTML file in any modern browser. The tool supports 8 widely-used brain atlases (Brainnetome 246, AAL-90/116, Schaefer 100/200/400, Power 264, and Dosenbach 160), provides real-time parameter adjustment through an intuitive graphical interface, and offers comprehensive edge management including click-to-connect, per-edge color customization, and Circos link file import. BrainRing supports both Chinese and English interfaces and enables researchers to produce publication-ready SVG and PNG figures with full control over visual styling, all within seconds rather than the minutes-to-hours workflow typical of script-based approaches. BrainRing is freely available at https://github.com/XiuFan719/brain-connectivity-viz with a live demo at https://XiuFan719.github.io/brain-connectivity-viz/.

cs.HC↗
arXiv2026

GeoTikzBridge: Advancing Multimodal Code Generation for Geometric Perception and Reasoning

Multimodal Large Language Models (MLLMs) have recently demonstrated remarkable perceptual and reasoning abilities. However, they struggle to perceive fine-grained geometric structures, constraining their ability of geometric understanding and visual reasoning. To address this, we propose GeoTikzBridge, a framework that enhances local geometric perception and visual reasoning through tikz-based code generation. Within this framework, we build two models supported by two complementary datasets. The GeoTikzBridge-Base model is trained on GeoTikz-Base dataset, the largest image-to-tikz dataset to date with 2.5M pairs (16 $\times$ larger than existing open-sourced datasets). This process is achieved via iterative data expansion and a localized geometric transformation strategy. Subsequently, GeoTikzBridge-Instruct is fine-tuned on GeoTikz-Instruct dataset which is the first instruction-augmented tikz dataset supporting visual reasoning. Extensive experimental results demonstrate that our models achieve state-of-the-art performance among open-sourced MLLMs. Furthermore, GeoTikzBridge models can serve as plug-and-play reasoning modules for any MLLM(LLM), enhancing reasoning performance in geometric problem-solving. Datasets and codes are publicly available at: https://github.com/sjy-1995/GeoTikzBridge.

cs.CV↗
arXiv2026

Adapting Frozen Mono-modal Backbones for Multi-modal Registration via Contrast-Agnostic Instance Optimization

Deformable image registration remains a central challenge in medical image analysis, particularly under multi-modal scenarios where intensity distributions vary significantly across scans. While deep learning methods provide efficient feed-forward predictions, they often fail to generalize robustly under distribution shifts at test time. A straightforward remedy is full network fine-tuning, yet for modern architectures such as Transformers or deep U-Nets, this adaptation is prohibitively expensive in both memory and runtime when operating in 3D. Meanwhile, the naive fine-tuning struggles more with potential degradation in performance in the existence of drastic domain shifts. In this work, we propose a registration framework that integrates a frozen pretrained \textbf{mono-modal} registration model with a lightweight adaptation pipeline for \textbf{multi-modal} image registration. Specifically, we employ style transfer based on contrast-agnostic representation generation and refinement modules to bridge modality and domain gaps with instance optimization at test time. This design is orthogonal to the choice of backbone mono-modal model, thus avoids the computational burden of full fine-tuning while retaining the flexibility to adapt to unseen domains. We evaluate our approach on the Learn2Reg 2025 LUMIR validation set and observe consistent improvements over the pretrained state-of-the-art mono-modal backbone. In particular, the method ranks second on the multi-modal subset, third on the out-of-domain subset, and achieves fourth place overall in Dice score. These results demonstrate that combining frozen mono-modal models with modality adaptation and lightweight instance optimization offers an effective and practical pathway toward robust multi-modal registration.

eess.IV↗
arXiv2026

Large language models for post-publication research evaluation: Evidence from expert recommendations and citation indicators

Assessing the quality of scientific research is essential for scholarly communication, yet widely used approaches face limitations in scalability, subjectivity, and time delay. Recent advances in large language models (LLMs) offer new opportunities for automated research evaluation based on textual content. This study examines whether LLMs can support post-publication peer review tasks by benchmarking their outputs against expert judgments and citation-based indicators. Two evaluation tasks are constructed using articles from the H1 Connect platform: identifying high-quality articles and performing finer-grained evaluation including article rating, merit classification, and expert style commenting. Multiple model families, including BERT models, general-purpose LLMs, and reasoning oriented LLMs, are evaluated under multiple learning strategies. Results show that LLMs perform well in coarse grained evaluation tasks, achieving accuracy above 0.8 in identifying highly recommended articles. However, performance decreases substantially in fine-grained rating tasks. Few-shot prompting improves performance over zero-shot settings, while supervised fine-tuning produces the strongest and most balanced results. Retrieval augmented prompting improves classification accuracy in some cases but does not consistently strengthen alignment with citation indicators. The overall correlations between model outputs and citation indicators remain positive but moderate.

cs.IR↗
arXiv2026

Breaking the degeneracy among regular black holes with gravitational lensing

We examine parameter degeneracies in Culetu, Bardeen and Hayward regular black holes across lensing, shadow and quasinormal mode regimes. Our analysis reveals that while Einstein ring data yield extremely loose constraints, with the regularization parameter $q$ exceeding $\mathcal{O}(10^3)$, they fail to improve the parameter estimation when combined with strong lensing observables. In contrast, the Event Horizon Telescope observations provide remarkably tight limits: $0 \leq q < 0.0466 <0.0847$ for Culetu, $0 \leq q < 0.5115 <0.6682$ for Bardeen and $0 \leq q < 1.0258 <1.1881$ for Hayward, which shows that the strong field regime alone dominates the available parameter space. Despite these bounds, leading order geometric observables remain highly degenerate, which masks the microscopic details of non-singular cores. To break this ``macroscopic universality,'' we identify high order signatures, such as the Lyapunov exponent and subleading time delays, as sensitive probes of near horizon curvature. Crucially, we discover that the brightness hierarchy of accretion induced intensity profiles undergoes a fundamental inversion when transitioning from lensing dominated static flows to dynamics dominated infalling flows. These results demonstrate that high resolution temporal and intensity profiles are essential for distinguishing between regular black hole geometries.

gr-qc↗
arXiv2026

TF4CTR: Twin Focus Framework for CTR Prediction via Adaptive Sample Differentiation

Effective feature interaction modeling is critical for enhancing the accuracy of click-through rate (CTR) prediction in industrial recommender systems. Most of the current deep CTR models resort to building complex network architectures to better capture intricate feature interactions or user behaviors. However, we identify two limitations in these models: (1) the samples given to the model are undifferentiated, which may lead the model to learn a larger number of easy samples in a single-minded manner while ignoring a smaller number of hard samples, thus reducing the model's generalization ability; (2) differentiated feature interaction encoders are designed to capture different interactions information but receive consistent supervision signals, thereby limiting the effectiveness of the encoder. To bridge the identified gaps, this paper introduces a novel CTR prediction framework by integrating the plug-and-play Twin Focus (TF) Loss, Sample Selection Embedding Module (SSEM), and Dynamic Fusion Module (DFM), named the Twin Focus Framework for CTR (TF4CTR). Specifically, the framework employs the SSEM at the bottom of the model to differentiate between samples, thereby assigning a more suitable encoder for each sample. Meanwhile, the TF Loss provides tailored supervision signals to both simple and complex encoders. Moreover, the DFM dynamically fuses the feature interaction information captured by the encoders, resulting in more accurate predictions. Experiments on five real-world datasets confirm the effectiveness and compatibility of the framework, demonstrating its capacity to enhance various representative baselines in a model-agnostic manner. To facilitate reproducible research, our open-sourced code and detailed running logs will be made available at: https://github.com/salmon1802/TF4CTR.

cs.IR↗
arXiv2026

CatBOX: A Categorical-Continuous Bayesian Optimization with Spectral Mixture Kernels for Accelerated Catalysis Experiments

Identifying optimal catalyst compositions and reaction conditions is central in catalysis research, yet remains challenging due to the vast multidimensional design spaces encompassing both continuous and categorical parameters. In this work, we present CatBOX, a Bayesian Optimization method for accelerated catalytic experimental design that jointly optimizes categorical and continuous experimental parameters. Our approach introduces a novel spectral mixture kernel that combines the inverse Fourier transform of Gaussian and Cauchy mixtures to provide a flexible representation of the continuous parameter space, capturing both smooth and non-smooth variations. Categorical choices, such as catalyst compositions and support types, are navigated via trust regions based on Hamming distance. For performance evaluation, CatBOX was theoretically verified based on information theory and benchmarked on a series of synthetic functions, achieving more than a 3-fold improvement relative to the best-performing baseline and a 19-fold improvement relative to random search on average across tasks. Additionally, three real-life catalytic experiments, including oxidative coupling of methane, urea-selective catalytic reduction, and direct arylation of imidazoles, were further used for in silico benchmarking, where CatBOX reliably identified top catalyst recipes and reaction conditions with the highest efficiencies in the absence of any a priori knowledge. Finally, we develop an open-source, code-free online platform to facilitate trial deployment in real experimental settings, particularly for self-driving laboratory environments.

cs.LG↗
arXiv2026

Improving Joint Audio-Video Generation with Cross-Modal Context Learning

The dual-stream transformer architecture-based joint audio-video generation method has become the dominant paradigm in current research. By incorporating pre-trained video diffusion models and audio diffusion models, along with a cross-modal interaction attention module, high-quality, temporally synchronized audio-video content can be generated with minimal training data. In this paper, we first revisit the dual-stream transformer paradigm and further analyze its limitations, including model manifold variations caused by the gating mechanism controlling cross-modal interactions, biases in multi-modal background regions introduced by cross-modal attention, and the inconsistencies in multi-modal classifier-free guidance (CFG) during training and inference, as well as conflicts between multiple conditions. To alleviate these issues, we propose Cross-Modal Context Learning (CCL), equipped with several carefully designed modules. Temporally Aligned RoPE and Partitioning (TARP) effectively enhances the temporal alignment between audio latent and video latent representations. The Learnable Context Tokens (LCT) and Dynamic Context Routing (DCR) in the Cross-Modal Context Attention (CCA) module provide stable unconditional anchors for cross-modal information, while dynamically routing based on different training tasks, further enhancing the model's convergence speed and generation quality. During inference, Unconditional Context Guidance (UCG) leverages the unconditional support provided by LCT to facilitate different forms of CFG, improving train-inference consistency and further alleviating conflicts. Through comprehensive evaluations, CCL achieves state-of-the-art performance compared with recent academic methods while requiring substantially fewer resources.

cs.CV↗
arXiv2026

UGID: Unified Graph Isomorphism for Debiasing Large Language Models

Large language models (LLMs) exhibit pronounced social biases. Output-level or data-optimization--based debiasing methods cannot fully resolve these biases, and many prior works have shown that biases are embedded in internal representations. We propose \underline{U}nified \underline{G}raph \underline{I}somorphism for \underline{D}ebiasing large language models (\textit{\textbf{UGID}}), an internal-representation--level debiasing framework for large language models that models the Transformer as a structured computational graph, where attention mechanisms define the routing edges of the graph and hidden states define the graph nodes. Specifically, debiasing is formulated as enforcing invariance of the graph structure across counterfactual inputs, with differences allowed only on sensitive attributes. \textit{\textbf{UGID}} jointly constrains attention routing and hidden representations in bias-sensitive regions, effectively preventing bias migration across architectural components. To achieve effective behavioral alignment without degrading general capabilities, we introduce a log-space constraint on sensitive logits and a selective anchor-based objective to preserve definitional semantics. Extensive experiments on large language models demonstrate that \textit{\textbf{UGID}} effectively reduces bias under both in-distribution and out-of-distribution settings, significantly reduces internal structural discrepancies, and preserves model safety and utility.

cs.CL↗
arXiv2026

Neighbor GRPO: Contrastive ODE Policy Optimization Aligns Flow Models

Group Relative Policy Optimization (GRPO) has shown promise in aligning image and video generative models with human preferences. However, applying it to modern flow matching models is challenging because of its deterministic sampling paradigm. Current methods address this issue by converting Ordinary Differential Equations (ODEs) to Stochastic Differential Equations (SDEs), which introduce stochasticity. However, this SDE-based GRPO suffers from issues of inefficient credit assignment and incompatibility with high-order solvers for fewer-step sampling. In this paper, we first reinterpret existing SDE-based GRPO methods from a distance optimization perspective, revealing their underlying mechanism as a form of contrastive learning. Based on this insight, we propose Neighbor GRPO, a novel alignment algorithm that completely bypasses the need for SDEs. Neighbor GRPO generates a diverse set of candidate trajectories by perturbing the initial noise conditions of the ODE and optimizes the model using a softmax distance-based surrogate leaping policy. We establish a theoretical connection between this distance-based objective and policy gradient optimization, rigorously integrating our approach into the GRPO framework. Our method fully preserves the advantages of deterministic ODE sampling, including efficiency and compatibility with high-order solvers. We further introduce symmetric anchor sampling for computational efficiency and group-wise quasi-norm reweighting to address reward flattening. Extensive experiments demonstrate that Neighbor GRPO significantly outperforms SDE-based counterparts in terms of training cost, convergence speed, and generation quality.

cs.CV↗
arXiv2026

Trust the Unreliability: Inward Backward Dynamic Unreliability Driven Coreset Selection for Medical Image Classification

Efficiently managing and utilizing large-scale medical imaging datasets with limited resources presents significant challenges. While coreset selection helps reduce computational costs, its effectiveness in medical data remains limited due to inherent complexity, such as large intra-class variation and high inter-class similarity. To address this, we revisit the training process and observe that neural networks consistently produce stable confidence predictions and better remember samples near class centers in training. However, concentrating on these samples may complicate the modeling of decision boundaries. Hence, we argue that the more unreliable samples are, in fact, the more informative in helping build the decision boundary. Based on this, we propose the Dynamic Unreliability-Driven Coreset Selection(DUCS) strategy. Specifically, we introduce an inward-backward unreliability assessment perspective: 1) Inward Self-Awareness: The model introspects its behavior by analyzing the evolution of confidence during training, thereby quantifying uncertainty of each sample. 2) Backward Memory Tracking: The model reflects on its training tracking by tracking the frequency of forgetting samples, thus evaluating its retention ability for each sample. Next, we select unreliable samples that exhibit substantial confidence fluctuations and are repeatedly forgotten during training. This selection process ensures that the chosen samples are near the decision boundary, thereby aiding the model in refining the boundary. Extensive experiments on public medical datasets demonstrate our superior performance compared to state-of-the-art(SOTA) methods, particularly at high compression rates.

cs.CV↗
arXiv2026

Transfer Learning for Contextual Joint Assortment-Pricing under Cross-Market Heterogeneity

We study transfer learning for contextual joint assortment-pricing under a multinomial logit choice model with bandit feedback. A seller operates across multiple related markets and observes only posted prices and realized purchases. While data from source markets can accelerate learning in a target market, cross-market differences in customer preferences may introduce systematic bias if pooled indiscriminately. We model heterogeneity through a structured utility shift, where markets share a common contextual utility structure but differ along a sparse set of latent preference coordinates. Building on this, we develop Transfer Joint Assortment-Pricing (TJAP), a bias-aware framework that combines aggregate-then-debias estimation with a UCB-style policy. TJAP constructs two-radius confidence bounds that separately capture statistical uncertainty and transfer-induced bias, uniformly over continuous prices. We establish matching minimax regret bounds of order $\tilde{O}\!\left(d\sqrt{\frac{T}{1+H}} + s_0\sqrt{T}\right),$revealing a transparent variance-bias tradeoff: transfer accelerates learning along shared preference directions, while heterogeneous components impose an irreducible adaptation cost. Numerical experiments corroborate the theory, showing that TJAP outperforms both target-only learning and naive pooling while remaining robust to cross-market differences.

stat.ME↗
arXiv2026

FaithSteer-BENCH: A Deployment-Aligned Stress-Testing Benchmark for Inference-Time Steering

Inference-time steering is widely regarded as a lightweight and parameter-free mechanism for controlling large language model (LLM) behavior, and prior work has often suggested that simple activation-level interventions can reliably induce targeted behavioral changes. However, such conclusions are typically drawn under relatively relaxed evaluation settings that overlook deployment constraints, capability trade-offs, and real-world robustness. We therefore introduce \textbf{FaithSteer-BENCH}, a stress-testing benchmark that evaluates steering methods at a fixed deployment-style operating point through three gate-wise criteria: controllability, utility preservation, and robustness. Across multiple models and representative steering approaches, we uncover several systematic failure modes that are largely obscured under standard evaluation, including illusory controllability, measurable cognitive tax on unrelated capabilities, and substantial brittleness under mild instruction-level perturbations, role prompts, encoding transformations, and data scarcity. Gate-wise benchmark results show that existing methods do not necessarily provide reliable controllability in deployment-oriented practical settings. In addition, mechanism-level diagnostics indicate that many steering methods induce prompt-conditional alignment rather than stable latent directional shifts, further explaining their fragility under stress. FaithSteer-BENCH therefore provides a unified benchmark and a clearer analytical lens for future method design, reliability evaluation, and deployment-oriented research in steering.

cs.AI↗