arXiv subjects

Jiawen Zhang

Publications and source records attributed to Jiawen Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

DirectorBench: Diagnosing Long-Form Video Generation with Personalized Multi-Agent Evaluation

Long-form video generation is rapidly moving from short, single-scene synthesis toward minute-long, multi-shot creation with narrative structure, cinematic control, audio, and cross-modal synchronization. However, evaluating such videos remains challenging, since existing benchmarks largely focus on local visual quality, short-horizon temporal consistency, or generic prompt alignment, and provide limited diagnosis of workflow failures and user-dependent preferences. We introduce DirectorBench, a personalized multi-agent diagnostic benchmark for long-form video generation. DirectorBench evaluates generated videos with respect to 80 structured metadata entries, 7 user profiles, and 40 checkpoint criteria across 5 dimensions: script, visual, audio, cross-modal, and stability. Instead of reducing quality to a single aggregate score, DirectorBench localizes checkpoint-level bottlenecks and supports profile-aware evaluation. We evaluate 4 long-form video generation workflows, 6 base LLMs, and 7 user profiles. Across workflows, DirectorBench reveals a between-unit bottleneck: transition quality averages only 0.256 and reaches 0.356 for the best workflow, while prompt-level user demand fulfillment averages 0.71. We further conduct human evaluation with 14 annotators to validate the alignment between DirectorBench and human judgment. The results show that DirectorBench captures human-perceptible quality differences and reveals workflow- and profile-dependent failure modes that are hidden by aggregate scoring. These findings highlight the importance of diagnostic and profile-aware benchmarking for long-form video generation.

cs.CL↗

arXiv2026

AssetGen: Deployable 3D Asset Generation at Interactive Speed

While 3D generation is progressing rapidly, recent work has often focused on obtaining high-resolution assets, leaving user experience and deployability as afterthoughts. We present AssetGen, a 3D generator that focuses instead on these two aspects. Given one reference image, in 30 seconds it produces a high-quality mesh with baked normals, a color texture, and a controlled polygon budget suitable for real-time rendering, including mobile use cases. The AssetGen Flash variant further reduces latency to 14 seconds for interactive and agentic creation loops. Our model generates the object geometry with a coarse-to-refine VecSet framework, which implements mesh simplification, cleaning, and normal baking on the GPU, and a fast parallel UV unwrapping. It then generates textures in a multi-view fashion, followed by backprojection and 3D inpainting. Model distillation, kernel optimization, and pipeline parallelization are co-designed to accelerate the system end-to-end. We introduce numerous automated and blind human evaluations and demonstrate competitive visual quality against leading commercial solutions in 30 seconds and preview-quality results in less than 15 seconds. The final result is a system that supports AI-assisted, deployable 3D content creation in interactive workflows.

cs.GR↗

arXiv2026

Evaluating Cognitive Age Alignment in Interactive AI Agents

While agentic AI and its core multimodal large language models (MLLMs) have demonstrated remarkable promise in language and visual reasoning across domains ranging from daily life to advanced scientific research, a profound gap remains between artificial and human intelligence. Despite the integration of powerful tools and advanced MLLMs, state-of-the-art AI agents frequently fail at foundational, seemingly simple tasks that a child can resolve with ease. Inspired by the Wechsler Intelligence Scale for Children (WISC), we introduce ChildAgentEval, the first psychometrically grounded interactive benchmark for evaluating cognitive age alignment in MLLM-based agents. ChildAgentEval systematically compares the reasoning performance of various MLLM-based interactive agents against age-specific human developmental stages, exposing where current agentic AI systems can and cannot simulate age-specific cognitive behavior.

cs.AI↗

arXiv2026

Global Regular Solutions of the Compressible Navier-Stokes Equations with Nonlinear Density-Dependent Viscosities and Large Initial Data of Spherical Symmetry

For the physically important case in which the viscosity coefficients depend on the density $ρ$ through a power law (i.e., $ρ^δ$ with some exponent $δ\in (\frac{1}{2},1)$), we establish the global well-posedness of regular solutions of the compressible Navier-Stokes equations for barotropic flow with large initial data of spherical symmetry in two and three spatial dimensions. The initial density considered here is positive everywhere but vanishes in the far field, ensuring that the resulting solutions satisfy the conservation laws of total mass and momentum. The most crucial step in our analysis is to obtain a uniform upper bound for the density, which is challenging due to the combined difficulties of degeneracy near the far-field vacuum, coordinate singularity at the origin, and nonlinearity of viscosity coefficients. Furthermore, the methodology developed here can also be applied to the corresponding problem in which the density remains strictly away from the vacuum.

math.AP↗

arXiv2026

SAID: Safety-Aware Intent Defense via Prefix Probing for Large Language Models

Large Language Models (LLMs) remain vulnerable to jailbreak attacks, where adversarially crafted prompts induce policy-violating responses despite safety alignment. Existing defenses typically improve safety through external filtering, auxiliary guardrails, or decoding-time control. However, these interventions often reduce practical deployability because they may require additional model access, introduce extra inference cost, or affect benign-task utility. In this paper, we propose Safety-Aware Intent Defense (SAID), a training-free jailbreak defense framework based on intent-level safety probing. SAID first distills potentially obfuscated user inputs into concise core intents using the target model itself. It then applies a validated safety prefix to probe each distilled intent and elicit the model's safety-aware response. Finally, a conservative aggregation rule rejects the original request if any distilled intent is identified as unsafe. This design enables black-box-compatible defense without updating model parameters or modifying the decoding process. Experiments on four open-source LLMs under six representative jailbreak attacks show that SAID achieves state-of-the-art defense performance in reducing harmful responses while maintaining competitive utility on benign tasks. Further analyses on prefix variants, hierarchical distillation, and inference efficiency demonstrate that SAID provides a practical safety-utility trade-off for securing LLMs against jailbreak threats.

cs.CR↗

arXiv2026

Confidence-Aware Alignment Makes Reasoning LLMs More Reliable

Large reasoning models often reach correct answers through flawed intermediate steps, creating a gap between final accuracy and reasoning reliability. Existing alignment strategies address this with external verifiers or massive sampling, limiting scalability. In this work, we introduce CASPO (Confidence-Aware Step-wise Preference Optimization), a framework that aligns token-level confidence with step-wise logical correctness through iterative Direct Preference Optimization, without training a separate reward model. During inference, we propose Confidence-aware Thought (CaT), which leverages this calibrated confidence to dynamically prune uncertain reasoning branches with negligible O(V) latency. Experiments across ten benchmarks and multiple model families show that CASPO consistently improves reasoning reliability and inference efficiency. CASPO scales to Qwen3-8B-Base and surpasses tree-search baselines on AIME'24 and AIME'25 without using reward-model data. We also release a step-wise dataset with confidence annotations to support fine-grained analysis of reasoning reliability. Code is available at https://github.com/Thecommonirin/CASPO.

cs.AI↗

arXiv2026

Mitigating Many-shot Jailbreak Attacks with One Single Demonstration

Many-shot jailbreaking (MSJ) causes safety-aligned language models to answer harmful queries by preceding them with many harmful question-answer demonstrations. We study why this attack becomes stronger as the number of demonstrations increases. Empirically, we find that MSJ induces a progressive activation drift: the representation of a fixed harmful query moves step by step away from the safety-aligned region as more harmful demonstrations are added. Theoretically, we show that this drift can be interpreted as implicit malicious fine-tuning: conditioning on N harmful demonstrations induces SGD-style updates equivalent to optimizing on the corresponding N harmful samples. This view turns the attack mechanism into a defense principle. We append a fixed one-shot safety demonstration at inference time, which induces a counteracting safety-oriented update and restores refusal behavior. The resulting method improves the model's robustness to MSJ without modifying its parameters or requiring white-box access at deployment. Code is available at https://github.com/Thecommonirin/SafeEnd.

cs.CR↗

arXiv2026

TGPP: Trajectory-Guided Plug-and-Play Priors for Sparse Radio Map Reconstruction

Radio map (RM) reconstruction is essential for environment-aware wireless networks, but practical measurements are often collected along mobility trajectories rather than randomly scattered over the target region. Such trajectory-sampled observations induce spatially heterogeneous uncertainty: near-trajectory regions are directly constrained, whereas distant or occluded regions remain weakly observed, leading to degraded reconstruction accuracy in under-constrained areas. To address this problem, we propose Trajectory-Guided Plug-and-Play Priors (TGPP), a general guidance module for sparse RM reconstruction. TGPP learns an explicit guidance map as an interpretable input-space risk prior, and an implicit guide feature that is projected and fused with backbone hidden representations. TGPP can be attached to different reconstruction backbones without changing their original task formulation. We further introduce RadioFlow-LDM, a latent flow-based generative backbone, and apply TGPP to deterministic, adversarial, graph-based, and latent generative reconstruction models. Experiments on RadioMapSeer with five trajectory sampling rates show that trajectory-sampled reconstruction differs substantially from random sparse interpolation. TGPP improves most reconstruction metrics across backbones, achieving up to 43.1% NMSE reduction relative to the corresponding base backbone without trajectory-guided priors.

eess.SP↗

arXiv2026

Global Regular Solutions of the Degenerate Compressible Navier-Stokes Equations with Large Initial Data of Spherical Symmetry

A fundamental open problem in the theory of the compressible Navier-Stokes equations is whether regular spherically symmetric flows can develop singularities, such as cavitation or implosion, in finite time. A formidable challenge lies in how the well-known coordinate singularity at the origin can be overcome to control the lower or upper bound of the density. In this paper, when the viscosity coefficients are degenerately density-dependent (as in the shallow water equations), we prove that, for general large spherically symmetric initial data with bounded positive density, solutions remain globally regular and cannot undergo cavitation or implosion in two and three spatial dimensions. Moreover, the far-field vacuum is allowed for the data under consideration here. Our results hold for all adiabatic exponents $γ\in(1,\infty)$ in two dimensions, and for physical adiabatic exponents $γ\in (1, 3)$ in three dimensions, without any restriction on the size of the initial data.

math.AP↗

arXiv2026

Tunable dislocations overcome mechano-functional tradeoff in perovskite oxides

Recent advancements in dislocation engineering are reshaping the traditional view towards ceramics being brittle. Here, we use KTaO3 (KTO), a perovskite oxide that is newly discovered with room-temperature bulk plasticity, and demonstrate that the seeded dislocations can effectively tune both mechanical and functional properties. We uncover a novel brittle-ductile-brittle (BDB) transition: low dislocation densities lead to brittle failure, intermediate densities (~10*14 m-2) enable superior ductility with strains over 20%, and high dislocation densities (~10*15 m-2) induce again brittle fracture. This dislocation density-dependent non-monotonic mechanical response challenges the traditional behavior of ceramics and offers new design opportunities. Furthermore, dislocation densities can monotonically decrease thermal conductivity, revealing a tradeoff between mechanical strength and functionality. The findings reveal a critical threshold of dislocation density in optimizing the performance of functional oxides, and provide a new framework for using dislocations to design advanced materials where mechanical durability and enhanced functionality are intertwined.

cond-mat.mtrl-sci↗

arXiv2026

Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution

We identify a critical security vulnerability in mainstream Claw personal AI agents: untrusted content encountered during heartbeat-driven background execution can silently pollute agent memory and subsequently influence user-facing behavior without the user's awareness. This vulnerability arises from an architectural design shared across the Claw ecosystem: heartbeat background execution runs in the same session as user-facing conversation, so content ingested from any external source monitored in the background (including email, message channels, news feeds, code repositories, and social platforms) can enter the same memory context used for foreground interaction, often with limited user visibility and without clear source provenance. We formalize this process as an Exposure (E) $\rightarrow$ Memory (M) $\rightarrow$ Behavior (B) pathway: misinformation encountered during heartbeat execution enters the agent's short-term session context, potentially gets written into long-term memory, and later shapes downstream user-facing behavior. We instantiate this pathway in an agent-native social setting using MissClaw, a controlled research replica of Moltbook. We find that (1) social credibility cues, especially perceived consensus, are the dominant driver of short-term behavioral influence, with misleading rates up to 61%; (2) routine memory-saving behavior can promote short-term pollution into durable long-term memory at rates up to 91%, with cross-session behavioral influence reaching 76%; (3) under naturalistic browsing with content dilution and context pruning, pollution still crosses session boundaries. Overall, prompt injection is not required: ordinary social misinformation is sufficient to silently shape agent memory and behavior under heartbeat-driven background execution.

cs.CR↗

arXiv2026

Non-collinear ferromagnetism in the Kondo lattice Ce$_5$CoGe$_2$

The dense Kondo lattice Ce$_5$CoGe$_2$ exhibits superconductivity once the magnetic ordering is suppressed by pressure. Here the ambient pressure magnetic state is investigated via magnetization, heat capacity, powder neutron diffraction, and muon spin relaxation ($μ$SR) measurements. Neutron diffraction results reveal a noncollinear ferromagnetic structure, where the four inequivalent Ce sites exhibit different magnetic moments. Point-charge model calculations of the crystalline-electric field (CEF) ground states corroborate different moments between the sites, and suggest sizeable components of the moments along different directions, consistent with the non-collinear structure. Analysis of the Dzyaloshinskii-Moriya (DM) interaction for the bonds connecting Ce atoms demonstrates that most of these bonds exhibit a nonzero DM vector, suggesting that competition between intersite magnetic exchange interactions, CEF driven single-ion anisotropy, the Kondo effect and the DM interaction may drive the non-collinear ferromagnetism.

cond-mat.str-el↗

arXiv2026

Probing a two-dimensional soft ferromagnet Cr$_2$Ge$_2$Te$_6$ by a tuning fork resonator

Magnetic anisotropy encodes key information about the free-energy landscape of magnetic materials, but its quantitative characterization often requires probes beyond conventional magnetometry. A quartz tuning-fork resonator provides direct access to the magnetotropic susceptibility. Here we use this technique to investigate the magnetic anisotropy of the layered ferromagnet Cr$_2$Ge$_2$Te$_6$. The temperature-, field-, and angle-dependent responses are consistently described by a quasi-two-dimensional (2D) easy-axis ferromagnetic model. In particular, the evolution of the magnetotropic susceptibility reveals how the angular profile changes from a conventional cos(2$θ$) form to a pronounced dip structure as the magnetization approaches directional saturation. These results establishCr$_2$Ge$_2$Te$_6$ as an ideal reference system for tuning-fork-based magnetotropic measurements. More broadly, they provide a useful framework for distinguishing spin-origin anisotropy from orbital magnetism, as in the case of CsV3Sb5. Our work demonstrates that tuning-fork resonators offer a sensitive thermodynamic probe of the rotational stiffness of magnetization in anisotropic low-dimensional magnets.

cond-mat.mtrl-sci↗

arXiv2026

From Pre-trained Models to Large Language Models: A Comprehensive Survey of AI-Driven Psychological Computing

The intersection of artificial intelligence and psychological science has experienced remarkable growth, with annual publications expanding from 859 papers in 2000 to 29,979 by 2025. However, this rapid evolution has created methodological fragmentation where similar computational techniques are independently developed across isolated psychological domains. This survey introduces the first systematic taxonomy that organizes AI-driven psychology tasks by computational processing patterns rather than application domains, categorizing them into four fundamental types: classification, regression, structured relational, and generative interactive tasks. Through analysis of over 300 representative works spanning the pre-trained model era and large language model era, we examine how computational approaches evolved from task-specific feature engineering to transfer learning and few-shot adaptation. We provide systematic coverage of datasets, evaluation metrics, and benchmarks while addressing fundamental challenges including interpretability, label uncertainty, privacy constraints, and cross-cultural validity. This computational perspective reveals transferable methodological patterns previously obscured by domain-centric organization, enabling systematic knowledge transfer and accelerated progress in computational psychology.

cs.CY↗

arXiv2026

Dimensionality tuning of heavy-fermion states in ultrathin CeSi2 films

Dimensionality tuning is an important method to modify the electronic states of quantum materials. However, the mechanism of such tuning in heavy fermion systems and its connection with transport properties remain largely unexplored. Here by combining molecular beam epitaxy (MBE), in-situ angle-resolved photoemission spectroscopy (ARPES) and transport measurements, we study the electronic states of the heavy-fermion compound CeSi2 as a function of film thickness. In three dimensional thick films, our measurements reveal a dispersive Kondo peak at the Fermi level (EF) and satellite peaks originating from crystal electric field (CEF) excitations, characteristic of heavy fermion systems. For two-dimensional ultrathin films, the CEF satellites are largely suppressed while the ground-state Kondo peak at EF remains strong, although it develops at lower temperatures. Simultaneously, the maximum temperature Tmax of the magnetic resistivity, \r{ho}m(T), changes from ~100 K in thick films to ~35 K in ultrathin films. This can be attributed to the dimensionality driven reduction of CEF excitations during the Kondo process, in good agreement with spectroscopic results. Our work provides direct insight to understand the quantum confinement effects on strongly correlated 4f-electron systems and opens up new opportunities to explore emergent phenomena in two-dimensional heavy-fermion materials.

cond-mat.str-el↗

arXiv2026

Pressure-induced superconductivity beyond magnetic quantum criticality in a Kondo ferromagnet

Quantum phase transitions are an established setting for emergent phenomena driven by strong electronic correlations, including strange metals and unconventional superconductivity. These have been explored extensively in Kondo lattice materials tuned to an antiferromagnetic quantum critical point (QCP), but superconductivity emerging near ferromagnetic quantum criticality is not yet observed, and the conditions under which it occurs in proximity to ferromagnetism are undetermined. Here, we report a new setting for superconductivity in the ferromagnetic Kondo-lattice material Ce5CoGe2, where there is a ferromagnetic ground state at ambient pressure, which evolves to antiferromagnetism under applied pressures. The antiferromagnetic transition is suppressed to a zero-temperature QCP, which is accompanied by strange-metal behavior. Superconductivity does not occur at the QCP, but instead appears at pressures beyond the magnetic instability. These findings suggest that Ce5CoGe2 represents a distinct class of correlated materials exhibiting a unique scenario for the emergence of superconductivity, likely associated with unconventional pairing mechanisms beyond spin-fluctuations.

cond-mat.supr-con↗

arXiv2026

Global Well-Posedness of the Vacuum Free Boundary Problem for the Degenerate Compressible Navier-Stokes Equations With Large Data of Spherical Symmetry

The study of global-in-time dynamics of vacuum is crucial for understanding viscous flows. In particular, physical vacuum, characterized by a moving boundary with nontrivial finite normal acceleration, naturally arises in the motion of shallow water. The corresponding large-data problems for multidimensional spherically symmetric flows remain open, due to the combined difficulties of coordinate singularity at the origin and degeneracy on the moving boundary. In this paper, we analyze the free boundary problem for the barotropic compressible Navier-Stokes equations with density-dependent viscosity coefficients (as in the shallow water equations) in two and three spatial dimensions. For a general class of spherically symmetric initial densities: $ρ_0^β\in H^3$ with $β\in (\frac{1}{3},γ-1]$ ($γ$: adiabatic exponent), vanishing on the moving boundary in the form of a distance function, we establish the global well-posedness of classical solutions with large initial data. We note that, when $β=γ-1$, $ρ_0$ contains a physical vacuum, but fails to satisfy the condition required for the Bresch-Desjardins (BD) entropy estimate when $γ\ge 2$, precluding the use of the BD entropy estimate to handle the degeneracy of the shallow water equations ({\it i.e.}, the case $γ=2$) on the physical vacuum boundary. Our analysis relies on a region-segmentation method: near the origin, we develop an interior BD entropy estimate, leading to flow-map-weighted estimates for the density; near the boundary, to handle the physical vacuum singularity, we introduce novel $ρ_0$-weighted estimates for the effective velocity, which are fundamentally different from the classical BD entropy estimate. Together, these estimates yield the desired global regularities.

math.AP↗

arXiv2026

Infinite Magnetoresistance and Vortex Coupling in the Pb/BSCCO Heterostructure

Combining superconductivity with spintronics provides exciting opportunities to realize low-dissipation quantum devices. Here we report the synthesis, characterization and magnetotransport measurements of the Pb/Bi$_2$Sr$_2$CaCu$_2$O$_{8+δ}$ (BSCCO) superconducting heterostructures, where an insulating PbO$_{x}$ layer spontaneously forms at the interface. Non-volatile switching between superconducting (logical "0") and normal ("1") states in Pb films by an external field, i.e., infinite magnetoresistance (IMR), can be realized and are attributed to the strong trapping and pinning of vortices in BSCCO. Furthermore, butterfly-shaped hysteresis loops in magnetoresistance, pronounced resistance dips/jumps and thermal reset to superconducting states can be observed and are direct manifestations of the peculiar vortex dynamics in BSCCO and vortex coupling across the Pb/BSCCO interface. Our work demonstrates a simple and effective way to realize IMR through superconducting vortices and opens up new opportunities to study the vortex interactions across the superconducting interfaces.

cond-mat.supr-con↗