arXiv subjects

Yue Zhou

Publications and source records attributed to Yue Zhou.

At least 37 records · Page 2Linked to original sources

arXiv2026

When Diffusion Models Forget Who You Are: Identity Preservation in Face Inpainting under Large Occlusions

Face inpainting with diffusion models has recently achieved impressive visual quality, yet preserving identity fidelity under significant occlusion and conflicting text guidance remains a major challenge. To address this issue, we present Reference Semantic Inpainting for Face (ReSem-Face), a cascaded diffusion framework that introduces an explicit identity-conditioned semantic prior for multi-reference face inpainting. Our approach distills representative identity features from multiple references to reconstruct missing semantic regions, which then guide the diffusion process through a multi-stream conditioning architecture. This design provides strong semantic constraints when pixels are absent and stabilizes identity reconstruction while remaining compatible with prompt-driven edits. Experiments on CelebAHQ-IDI-5 and VGGFace2 demonstrate that ReSem-Face yields more reliable identity-preserving completion under severe semantic masks and improves text-controlled editing quality compared with representative baselines.

cs.CV↗

arXiv2026

On the autotopism groups and the equivalence of finite cyclic semifields

Special cases of finite cyclic semifields were first constructed by Hughes and Kleinfeld in 1960, and later by Sandler in 1962 and Knuth in 1965. The general construction of cyclic semifields was subsequently introduced by Petit in 1966, and later rediscovered from the perspective of irreducible semilinear transformations by Jha and Johnson in 1989. Since Sandler's foundational work in 1962, the complete determination of the autotopism groups of cyclic semifields and the full resolution of the isotopy problem for this family have remained long-standing open problems. The most significant advances in determining these autotopism groups are due to Dempwolff in 2011, who left open the case in which the field extension degree strictly divides the degree of the polynomial defining the semifield. In this paper, we provide a complete classification of cyclic semifields up to isotopy, together with the full determination of their autotopism groups, thereby closing the remaining cases left open by Dempwolff. Since cyclic semifields arise as a special instance of a broader family of maximum rank distance (MRD) codes constructed via skew polynomials, our methods also yield a complete classification of these MRD codes up to linear and semilinear equivalence over the prime field, together with an explicit description of their full automorphism groups.

math.CO↗

arXiv2026

Three-dimensional evolution of a solar filament with multipoint observations

In this paper, we first devise a geometrical model, featuring a torus-like flux rope based on the shape of 3DCORE model. The global shape of the torus is an ellipse, while the cross sections are circular along the torus. The thinnest point is located between the Sun center and photosphere. Deflections and inclination are considered as well. Using multiwavelength observations from perspectives of Earth, Ahead-STEREO (STA), and Solar Orbiter, we apply the model to three-dimensional (3D) reconstructions and tracking of the filament eruption, which was associated with a flare and a coronal mass ejection (CME) on 2024 October 8. The morphology, direction, and true velocity ($\sim$433 km/s) of the eruptive filament are obtained. It is found that the filament propagates nonradially, deflecting slightly eastward by $\sim$10 degrees and significantly southward by $\sim$40 degrees. Trajectory of the filament in the ecliptic plane reveals that the filament moves toward STA. The true direction of the eruptive filament using imaging and spectral observations is mutually verified by 3D reconstructions. The heliocentric distance of the filament increases from $\sim$1.68 to $\sim$2.94 solar radii within 35 minutes. Based on the results of 3D reconstructions, the true speed of the CME leading edge is evaluated to be 1046$-$1145 km/s.

astro-ph.SR↗

arXiv2026

Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation

Scientific ideas rarely start from a blank page. They inherit mechanisms, repair known limitations, and recombine pieces of earlier work, much like biological genomes. Current benchmarks still say little about whether AI systems can follow this inheritance structure. We present IdeaGene-Bench (IG-Bench), a benchmark for scientific lineage reasoning and lineage-grounded idea generation. IG-Bench is organized around the IdeaGene framework: each paper or proposal is represented as a set of minimal, typed, evidence-grounded Idea Genome objects, and a GenomeDiff aligns these objects to record inheritance, mutation, loss, external import, and novel insertion under six operational evolutionary dynamics. The benchmark contains 1,961 golden lineage traces, 1,085 curated Idea Genome objects, and 920 pairwise GenomeDiff records across 10 scientific domains. It supports two evaluations. IG-Exam (42 task types, 1,029 instances) tests closed-form lineage reasoning across Idea Genome abstraction, inheritance tracing, evolutionary reasoning, and lineage verification. IG-Arena evaluates generation with a lineage-conditioned Population-Evolution Score(PES), asking whether a proposal can be inserted as a coherent descendant of a given lineage population: it should inherit the right Idea Genome objects, vary meaningfully from nearby work, and offer selection value for future research. Experiments on 14 LLM-based scientists expose a compositional bottleneck. The strongest system reaches only 27.3% exact accuracy on lineage reasoning, and structured lineage context reshuffles system rankings rather than helping every participant uniformly.

cs.AI↗

arXiv2026

Boundedness and blow-up for a quasilinear Keller-Segel system with flux limitation and indirect signal production

The quasilinear Keller-Segel system with flux limitation and indirect signal production u_t=\nabla\cdot\left(D(u)\nabla u\right) -\nabla\cdot\left(u(1+\left|\nabla v\right|^2)^σ\nabla v\right), &x\inΩ, t>0, \\ 0=Δv-v+w,x\inΩ, t>0, \\ w_t=-w+u,x\inΩ, t>0, under homogeneous Neumann boundary conditions in a smooth bounded domain Ω\subsetR^N is considered, where D(u)\simeq u^{m-1} as u\simeq\infty. We conclude that For N=1 and any σ\in\mathbb{R}, if m\geq0, the classical solution exists globally, and it is moreover bounded if m>0. However, if m<0 and Ωis a ball, there exist radially symmetric initial data such that the classical solution exhibits finite-time blow-up. For any N\geq2 and m>1-\frac{1}{N}, if σ\leq \frac{mN+2-2N}{2N-2}, the classical solution is global. Furthermore, if σ<\frac{mN+2-2N}{2N-2}, the corresponding solution is uniformly bounded. For any N\geq2 and m<2-\frac{2}{N}, if σ>\max\left\{\frac{N}{2-2N},\frac{mN+2-2N}{2N-2}\right\} and Ωis a ball, there exist radially symmetric initial data such that the classical solution blows up in finite time.

math.AP↗

arXiv2026

AgentFoX: LLM Agent-Guided Fusion with eXplainability for AI-Generated Image Detection

The realism of AI-generated images (AIGI) poses increasing challenges for reliable forensic detection, where heterogeneous expert detectors may produce conflicting predictions across diverse generative sources and post-processing conditions. Existing multi-expert fusion methods rely on fixed rules or learned fusion strategies, offering limited ability to assess sample-specific reliability, execute rigorous adjudication of conflicts, and provide evidence-grounded explanations. We propose AgentFoX, an LLM-driven agentic multi-expert framework for AIGI detection that employs a command-and-reasoning core to perform evidence fusion. Following predefined guidelines, the core coordinates designated subtasks to collect semantic and signal-level evidence, reason over structured contexts to determine authenticity, and generate an auditable report for explainability. During this process, Expert Profiles are constructed for model-centric reliability assessment, while Clustering Profiles are built for data-centric contextual analysis, jointly establishing evidence contexts for conflict resolution. Extensive evaluations across diverse benchmarks demonstrate the robustness and generalizability of AgentFoX under complex conditions.

cs.CV↗

arXiv2026

Deep Residual Injection for Full-Spectrum Forensic Signal Perception in Multimodal Large Language Models

Multimodal large language models (MLLMs) have been increasingly adopted in forensics for their robust semantic understanding. As AI-generated images become realistic, semantic-level inconsistencies alone are often insufficient for reliable detection. This motivates a critical question: whether MLLMs can achieve full-spectrum forensic signal perception, i.e., capturing low-level generator artifacts without sacrificing pre-trained semantic knowledge. We further perform a layer-wise analysis of forensic signal perception in MLLMs, showing that semantic information is primarily formed in the early-to-middle layers, whereas direct fine-tuning for artifact learning disrupts these semantic representations. Based on this insight, we propose Deep Visual Residual MLLM (Deep-VRM) to preserve early semantic processing while injecting artifact-specific visual signals as a residual path into an intermediate layer, where they are fused with semantic token representations and propagated through subsequent trainable layers. This enables later layers to jointly model semantic reasoning and signal-level forensic cues, and surprisingly, the model learns to adaptively leverage different levels of forensic signals depending on the input, achieving robust and generalizable detection performance. Extensive experiments show that our method achieves state-of-the-art across most benchmarks. The code and data are available at https://github.com/KQL11/Deep-VRM.

cs.CV↗

arXiv2026

Scalable Event Cloud Network for Event-based Classification

Event cameras are biologically inspired sensors garnering significant attention from both industry and academia. Mainstream methods favor frame and voxel representations, which reach a satisfactory performance while introducing time-consuming transformations, bulky models, and sacrificing fine-grained temporal information. Alternatively, Point Cloud representation demonstrates promise in addressing the mentioned weaknesses, but it has limited scalability in abstracting features of higher spatial resolution and longer temporal sequence events. In this paper, we propose a Scalable Network named SECNet to leverage Event Cloud representation. SECNet integrates polarity at the structural level by innovating the Event-based Group and Sampling module rather than only at the input level. To accommodate the surge in the number of events, SECNet embraces feature extraction in the frequency domain via the Fourier transform.This approach not only substantially extinguishes the explosion of Multiply Accumulate Operations but also effectively abstracts spatio-temporal features. We conducted extensive experiments on \textbf{ten} event-based datasets, and substantiate the scalability, effectiveness, and efficiency of SECNet. Our code will be available at: https://github.com/rhwxmx/SECNet_ICML.

cs.CV↗

arXiv2026

What Do LLMs Know About Alzheimer's Disease? Multi-loss Fine-Tuning and Probing for AD Detection

Reliable early detection of Alzheimer's disease (AD) is challenging, particularly due to the limited availability of labeled data. While large language models (LLMs) have shown strong transfer capabilities across do mains, adapting them to the AD domain through supervised fine-tuning remains largely unexplored. In this work, we empirically evaluate various model architectures across three heterogeneous transcript corpora (Pitt, CCC, ADRC) to investigate their effectiveness for text-based AD detection and analyze how task-relevant information is encoded within their internal representations. To the best of our knowledge, our fine-tuned BERT and T5 models establish a new state-of-the-art on the Pitt and CCC datasets, while achieving strong performance on ADRC. In parallel, the decoder-only Llama-1B achieves highly competitive results comparable to BERT and T5 across all three corpora, highlighting its effectiveness for AD detection. We further conduct a comprehensive evaluation of the Llama-1B backbone, analyzing cross-corpus transferability, optimal input chunk-size granularity, and the impact of clinical transcript markers. Also, we use linear probing to empirically show that fine-tuning shifts the representations of individual tokens, both linguistic markers and content words, in ways that reflect AD-related signal.

cs.CL↗

arXiv2026

From Accounting to Coordination: A Virtual Water-Aware Electricity-Computation-Water Nexus Framework for Data Center Dispatch

The expansion of data centers (DCs) drives a sustained increase in electricity demand and associated water withdrawals at generation sites. These withdrawals occur at generation sites and are virtually allocated to demand based on network power flows. Consequently, the actual water footprint of a specific load varies dynamically with generation dispatch and network conditions. Existing approaches typically rely on static statistical accounting to quantify these water footprints. However, such static methods fail to capture how dispatch optimization and workload relocation dynamically affect water withdrawals. As a result, static statistical accounting approaches remain decoupled from the optimization process, rendering them incapable of guiding workload relocation or power dispatch to mitigate water stress. To address this limitation, this paper develops an operational electricity-computation-water (ECW) nexus framework that internalizes virtual water impacts directly into power system dispatch. The framework represents dispatch optimization as a differentiable optimization layer embedded within a deep learning architecture, enabling efficient end-to-end learning of coordination policies while preserving operational feasibility. Combined with fixed-point coordination, the framework enforces consistency between virtual water attribution and physical generation-side withdrawals. Case studies on the IEEE 30-bus and 118-bus test systems demonstrate reliable convergence, exact power-water consistency, and reductions of approximately 3-5% in generation-related freshwater withdrawals under water-constrained conditions.

cs.AI↗

arXiv2026

Look-Closer-Then-Diagnose: Confidence-Aware Ultrasound VQA via Active Zooming

Vision-Language Models (VLMs) have significantly advanced medical visual question answering, yet their performance in ultrasound remains suboptimal. In clinical practice, sonographers explicitly focus on lesion regions to formulate reports, though diagnostic interpretations sometimes vary due to inherent subjectivity. However, existing VLMs are not explicitly structured to interactively zoom into lesions prior to diagnosis; moreover, they typically treat annotations as unbiased ground truths, failing to account for their inherent subjectivity and ambiguity. In this paper, we propose a framework specifically designed to consider the sonographer's cognitive workflow. We first introduce a structured Zoom-then-Diagnose paradigm, which replicates the interactive search process to enable lesion-focused reasoning. Furthermore, within the Group Relative Policy Optimization (GRPO) framework, we introduce an uncertainty-aware reward derived from stochastic group-wise rollouts to estimate prediction consistency as a proxy for model confidence. Together, these two components encourage the model to reinforce accurate predictions on clear cases while remaining cautious under ambiguity. Experiments across liver, breast, and thyroid datasets show that our framework improves lesion localization by 39.3\%, demonstrating that our model has learned the ability to actively look closer and diagnose.

cs.CV↗

arXiv2026

A generalization of Kadell's orthogonality ex-conjecture

In 2000, Kadell gave an orthogonality conjecture for a symmetric function generalization of the Zeilberger--Bressoud $q$-Dyson constant term identity. The non-zero part of Kadell's conjecture is a constant term identity indexed by a weak composition $v$. This conjecture was first proved by Károlyi, Lascoux and Warnaar in 2015. They further formulated a closed-form expression for the above constant term when all parts of the composition $v$ are distinct. In 2021, Zhou obtained a recursion for this constant term for an arbitrary composition $v$. In this paper, by categorizing the variables into two parts, we generalize Zhou's result.

math.CO↗

arXiv2026

Resolving Action Bottleneck: Agentic Reinforcement Learning Informed by Token-Level Energy

Agentic reinforcement learning trains large language models using multi-turn trajectories that interleave long reasoning traces with short environment-facing actions. Common policy-gradient methods, such as PPO and GRPO, treat each token in a trajectory equally, leading to uniform credit assignment. In this paper, we critically demonstrate that such uniform credit assignment largely misallocates token-level training signals. From an energy-based modeling perspective, we show that token-level training signals, quantified by their correlations with reward variance of different rollouts sampled from a given prompt, concentrate sharply on action tokens rather than reasoning tokens, even though action tokens account for only a small fraction of the trajectory. We refer to this phenomenon as the Action Bottleneck. Motivated by this observation, we propose an embarrassingly simple token reweighting approach, ActFocus, that downweights gradients on reasoning tokens, along with an additional energy-based redistribution mechanism that further increases the weights on action tokens with higher uncertainty. Across four environments and different model sizes, ActFocus consistently outperforms PPO and GRPO, yielding final-step gains of up to 65.2 and 63.7 percentage points, respectively, without any additional runtime or memory cost.

cs.LG↗

arXiv2026

On the dimension of the space generated by characteristic vectors of $q$-Steiner systems

Fix a prime power $q$ and parameters $1\leq t\leq k\leq n$, the corresponding Steiner system in the Grassmann scheme, or the $q$-Steiner system, is a collection $\mathfrak{B}$ of $k$-dimensional subspaces of $\mathbb{F}_{q}^n$ such that for each $t$-dimensional subspace $T$, there exists exactly one element of $\mathfrak{B}$ containing $T$. The dimension of Steiner systems in the Grassmann scheme is defined to be the dimension of the $\mathbb{Q}$-vector space spanned by the characteristic vectors of all these $q$-Steiner systems. In this paper, we prove that when a quadruple $(t,k,n,q)$ admits at least one $q$-Steiner system, the corresponding dimension is equal to ${n\brack k}_{q}-{n\brack t}_{q}+1$. This generalizes the 2019 work of Ghodrati \cite{ghodrati2019dimension} on ordinary Steiner systems.

math.CO↗

arXiv2026

LLM-Based Human-Agent Collaboration and Interaction Systems: A Survey

Recent advances in large language models (LLMs) have sparked growing interest in building fully autonomous agents. However, fully autonomous LLM-based agents still face significant challenges, including limited reliability due to hallucinations, difficulty in handling complex tasks, and substantial safety and ethical risks, all of which limit their feasibility and trustworthiness in real-world applications. To overcome these limitations, LLM-based human-agent systems (LLM-HAS) incorporate human-provided information, feedback, or control into the agent system to enhance system performance, reliability, and safety. These human-agent collaboration systems enable humans and LLM-based agents to collaborate effectively by leveraging their complementary strengths. This paper provides the first comprehensive and structured survey of LLM-HAS. It clarifies fundamental concepts, systematically presents core components shaping these systems, including environment and profiling, human feedback, interaction types, orchestration, and communication, explores emerging applications, and discusses unique challenges and opportunities arising from human-AI collaboration. By consolidating current knowledge and offering a structured overview, we aim to foster further research and innovation in this rapidly evolving interdisciplinary field. Paper lists and resources are available at https://github.com/HenryPengZou/Awesome-Human-Agent-Collaboration-Interaction-Systems.

cs.CL↗

arXiv2026

Early Risk Prediction with Temporally and Contextually Grounded Clinical Language Processing

Clinical notes in Electronic Health Records (EHRs) capture rich temporal information on events, clinician reasoning, and lifestyle factors often missing from structured data. Leveraging them for predictive modeling can be impactful for timely identification of chronic diseases. However, they present core natural language processing (NLP) challenges: long text, irregular event distribution, complex temporal dependencies, privacy constraints, and resource limitations. We present two complementary methods for temporally and contextually grounded risk prediction from longitudinal notes. First, we introduce HiTGNN, a hierarchical temporal graph neural network that integrates intra-note temporal event structures, inter-visit dynamics, and medical knowledge to model patient trajectories with fine-grained temporal granularity. Second, we propose ReVeAL, a lightweight test-time framework that distills LLMs' reasoning into smaller verifier models. Applied to opportunistic screening for Type 2 Diabetes (T2D) using temporally realistic cohorts curated from private and public hospital corpora, HiTGNN achieves the highest predictive accuracy, especially for near-term risk, while preserving privacy and limiting reliance on large proprietary models. ReVeAL enhances sensitivity to true T2D cases and retains explanatory reasoning. Our ablations confirm the value of temporal structure and knowledge augmentation, and fairness analysis shows HiTGNN performs more equitably across subgroups.

cs.CL↗

arXiv2026

A Process-Aware Demand Response Evaluation Framework for Hydrogen-Integrated Zero-Carbon Steel Plants Coupled with Methanol Production

High penetration of renewables (RES) and the retirement of thermal units aggravate flexibility scarcity in power systems. Hydrogen-based low-carbon steel production systems possess substantial demand response (DR) potential. This paper proposes a process-aware DR evaluation framework for hydrogen-integrated zero-carbon steel plants coupled with methanol production (H2-DRI-EAF-MeOH). First, a novel H2-DRI-EAF-MeOH architecture is introduced to eliminate residual emissions via methanol synthesis. Integrated energy-material flows are formulated to reflect coupling interactions governing DR potential. Second, to capture electric arc furnace (EAF) operational constraints while preserving tractability, an operating feasible region model is developed and validated using field data from a pure hydrogen direct reduced iron and EAF plant, yielding a 4.1% average relative error. Third, a process-aware DR potential evaluation model is formulated, incorporating a nonlinear asymmetric penalty and an adaptive rolling mechanism to reflect operators' aversion to process deviations and avoid myopic scheduling. Finally, dual-side evaluation metrics are established to quantify grid-side delivered DR capacity and ramping risks, making load-side unit-level regulation behaviors observable. Case studies show the proposed framework achieves an average effective delivered DR capacity of 178.3 MW, improves RES-load matching from 0.257 to 0.587, and reduces costs by 15.68% compared to the baseline. Furthermore, the exponential asymmetric penalty mitigates extreme tail risks of process deviations. Ultimately, this work provides a theoretical foundation for leveraging RES-steel-chemical synergies to mitigate flexibility scarcity.

eess.SY↗

arXiv2026

Simplicity Prevails: The Emergence of Generalizable AIGI Detection in Visual Foundation Models

While specialized detectors for AI-Generated Images (AIGI) achieve near-perfect accuracy on curated benchmarks, they suffer from a dramatic performance collapse in realistic, in-the-wild scenarios. In this work, we demonstrate that simplicity prevails over complex architectural designs. A simple linear classifier trained on the frozen features of modern Vision Foundation Models , including Perception Encoder, MetaCLIP 2, and DINOv3, establishes a new state-of-the-art. Through a comprehensive evaluation spanning traditional benchmarks, unseen generators, and challenging in-the-wild distributions, we show that this baseline not only matches specialized detectors on standard benchmarks but also decisively outperforms them on in-the-wild datasets, boosting accuracy by striking margins of over 30\%. We posit that this superior capability is an emergent property driven by the massive scale of pre-training data containing synthetic content. We trace the source of this capability to two distinct manifestations of data exposure: Vision-Language Models internalize an explicit semantic concept of forgery, while Self-Supervised Learning models implicitly acquire discriminative forensic features from the pretraining data. However, we also reveal persistent limitations: these models suffer from performance degradation under recapture and transmission, remain blind to VAE reconstruction and localized editing. We conclude by advocating for a paradigm shift in AI forensics, moving from overfitting on static benchmarks to harnessing the evolving world knowledge of foundation models for real-world reliability.

cs.CV↗