arXiv subjects

Yang Zhang

Publications and source records attributed to Yang Zhang.

At least 163 records · Page 9Linked to original sources

arXiv2026

LongSumEval: Question-Answering Based Evaluation and Feedback-Driven Refinement for Long Document Summarization

Evaluating long document summaries remains the primary bottleneck in summarization research. Existing metrics correlate weakly with human judgments and produce aggregate scores without explaining deficiencies or guiding improvement, preventing effective refinement in applications requiring verifiable accuracy. We introduce LongSumEval, a unified framework bridging evaluation and generation through structured question-answering feedback. The framework operationalizes summary quality as answerability and factual alignment of question-answer pairs, generating interpretable scores and actionable feedback that identifies coverage gaps and factual inconsistencies. This resolves the misalignment where evaluation operates independently of generation objectives. Meta-evaluation of our QA-based evaluation module across seven benchmarks demonstrates substantially stronger agreement with human judgments compared to established metrics. Structured feedback enables significant quality improvements through self-refinement without retraining. By demonstrating that evaluation feedback can serve as executable instructions for generation, this work establishes a generalizable paradigm for aligning assessment with improvement, with direct implications for controllable text generation requiring verifiable accuracy and transparent quality control. All code and datasets will be released in GitHub for reproducibility.

cs.CL↗

arXiv2026

LLM-ReSum: A Framework for LLM Reflective Summarization through Self-Evaluation

Reliable evaluation of large language model (LLM)-generated summaries remains an open challenge, particularly across heterogeneous domains and document lengths. We conduct a comprehensive meta-evaluation of 14 automatic summarization metrics and LLM-based evaluators across seven datasets spanning five domains, covering documents from short news articles to long scientific, governmental, and legal texts (2K-27K words) with over 1,500 human-annotated summaries. Our results show that traditional lexical overlap metrics (e.g., ROUGE, BLEU) exhibit weak or negative correlation with human judgments, while task-specific neural metrics and LLM-based evaluators achieve substantially higher alignment, especially for linguistic quality assessment. Leveraging these findings, we propose LLM-ReSum, a self-reflective summarization framework that integrates LLM-based evaluation and generation in a closed feedback loop without model finetuning. Across three domains, LLM-ReSum improves low-quality summaries by up to 33% in factual accuracy and 39% in coverage, with human evaluators preferring refined summaries in 89% of cases. We additionally introduce PatentSumEval, a new human-annotated benchmark for legal document summarization comprising 180 expert-evaluated summaries. All code and datasets will be released in GitHub.

cs.CL↗

arXiv2026

Compact Spin-Charge Separated Neural Quantum States for Valence-Bond States

Neural-network quantum states (NQS) provide a flexible nonlinear representation of quantum many-body wavefunctions, but their efficiency depends sensitively on whether the architecture reflects the sign structure and constrained Hilbert space of the target state. In this work, we propose a solvable-point-guided strategy: design the architecture at an exactly solvable point where the correct local rules can be read off, then refine to the non-exact regime by enlarging only the kernel size and hidden dimension. The strategy is built from four physics-motivated designs: a stride-matched local-rule convolution, geometric pooling, a sign-resolving $\tanh(x^{2k+1})$ activation, and explicit spin-hole sector separation. We test this approach on quasi-one-dimensional valence-bond-solid (VBS) states and their doped soliton variants (sVBS), the exact ground states of a $t$-$J$-like model with a single mobile hole. In finite-size benchmarks, this architecture reaches high fidelity for the exact sVBS state with substantially fewer parameters than generic fully connected, convolutional, and transformer baselines tested under the same setup. For the spin sector, the learned local rule transfers from small to larger systems without retraining. Away from the solvable point, increasing kernel size and hidden dimension systematically improves accuracy, and the model shows approximately $L^2$ parameter scaling in the gapless regime for system size $L$, compared with approximately $L^4$ for matrix-product states in the same regime. Our work establishes a recipe for compact NQS in sign-structured, constrained Hilbert spaces and paves the pathway to physics-informed architectures for the broader $t$-$J$ and Hubbard families.

cond-mat.str-el↗

arXiv2026

Time-Unconditional Generative Speech Enhancement via Autonomous Rectified Flow

Most generative speech enhancement methods rely on explicit time-step embeddings for temporal conditioning. In this paper, we propose the Autonomous Rectified Flow framework, which challenges the necessity of such conditioning. Using a linear interpolation path, we show that the target vector field is inherently time-invariant. We further introduce a time-unconditional network that eliminates explicit time-step information and infers the denoising direction solely from the spatial relationship between the current state and the noisy observation. Predicting this target vector field is equivalent to modeling the noise distribution. By avoiding overfitting to temporal trajectories, the proposed autonomous design significantly improves generation quality, robustness, and inference efficiency.

eess.AS↗

arXiv2026

When Safe Concepts Become Unsafe: Multi-Concept Compositional Vulnerabilities in Text-to-Image Models

Text-to-image (T2I) models are increasingly optimized for following user instructions faithfully. However, we find that this capability introduces a safety vulnerability we call Multi-Concept Compositional Unsafety (MCCU). MCCU occurs when multiple individually safe concepts, if combined in a single generation request, lead to harmful or sensitive visual outputs. Unlike prior jailbreak settings, MCCU does not rely on adversarial prompts, model access, or explicitly disallowed content. Instead, the risk emerges from how the model composes multiple safe visual concepts into a single scene. To systematically measure this threat, we build TwoHamsters, a large-scale evaluation framework consisting of 20k prompts, 51 curated concept pairs, and six risk categories. We evaluate 13 T2I models under a black-box setting. Our results show a clear conflict between instruction-following and safety: models that follow prompts more faithfully tend to produce more MCCU failures. For example, FLUX.1 achieves a 99.35% Unsafe Alignment Rate while only reaching a 1.57% MCCU Defense Rate. We further evaluate three representative defenses, including safety filtering, MCCU-specific detector fine-tuning, and concept erasure, all of which fail against unseen concept combinations. Our findings suggest that compositional reasoning in T2I models creates an attack surface that is not captured by existing safety mechanisms. We anticipate the release of TwoHamsters will catalyze community development of advanced generative defense mechanisms.

cs.CR↗

arXiv2026

Plasma Flow Generation and Particle Acceleration from Expanding Magnetic Bubbles

Impulsive plasma dynamics in the laboratory are often driven by rising electric currents, yet their quantitative plasma response has not been well established. By means of fully kinetic particle-in-cell simulations and laser-driven capacitor-coil experiments, we show that a rising current expels plasma, forming an expanding magnetic bubble and accelerating particles. The expansion front velocity scales with the Alfvén speed determined by the magnetic field at its inner edge and the plasma density at its outer edge. This mechanism establishes impulsive current drive as a fundamental way that generates plasma flows and accelerates particles in laboratory plasmas, with potential relevance to astrophysics.

physics.plasm-ph↗

arXiv2026

ScanBot: A Benchmark for Precision Robotic Surface Scanning with Industrial Laser Profilers

We introduce ScanBot, a benchmark for instruction-conditioned, high-precision surface scanning with robot-mounted industrial laser profilers. Unlike existing robot learning datasets that emphasize coarse behaviors such as grasping, navigation, or dialogue, ScanBot targets sensing-centric tasks where sub-millimeter motion continuity, strict stand-off control, and stable scanner settings are essential for acquiring usable geometry. The dataset contains scanning trajectories over twenty objects, including electronic components and structured 3D-printed parts, and spans six task types that range from broad inspection to fine-grained detail scanning and geometry-critical operations, including metrology and registration. Each episode is specified by natural language instructions and paired with synchronized first-person RGB-D, third-person video, laser height profiles, robot joint and pose traces, and scanner-parameter logs. These requirements expose a gap: despite recent progress, learning-based models often fail to produce stable and feasible scan motions under fine-grained instructions and real laser-profiling constraints. To reflect how industrial scanning is actually done, we evaluate methods through a two-stage pipeline. Stage I asks the model to "set up the sensor" by recommending scanner parameters, while Stage II asks it to "move like a scanner" by producing smooth, feasible trajectories that maintain stand-off and cover the intended region under precision demands.

cs.RO↗

arXiv2026

How Should a Robot Configure Its Laser Scanner for Inspection?

Robotic inspection relies on accurate sensing to acquire high-fidelity geometric measurements for defect detection and metrology. While prior work has focused on robot motion and viewpoint planning, how to configure sensing parameters remains largely underexplored, despite their decisive impact on measurement quality. We propose SenseHD, a robotic sensing system that formulates scanner configuration as an instruction-conditioned sensing decision. Instead of predicting precise parameter values, SenseHD treats sensing parameters as discrete sensing actions and selects stable sensing regimes through hyperdimensional associative memory. Experiments on a real robotic inspection platform demonstrate that SenseHD robustly selects appropriate sensing configurations and significantly improves inspection reliability, while remaining lightweight and efficient compared to baseline methods.

cs.RO↗

arXiv2026

Flow mechanisms governing oscillation in a sonic fluidic oscillator

Two factors that influence the oscillation mechanism of a sonic fluidic oscillator are investigated: the geometry of the feedback channel connections (control ports) and the influence of flow restrictions in the oscillator outlets. Phase-averaged planar PIV measurements are performed inside the oscillator, synchronised with unsteady pressure measurements, and analysed using space-only proper orthogonal decomposition (POD). The POD analysis reveals two coupled modes: a Sweeping Mode capturing lateral jet displacement and a Bending Mode capturing jet curvature during switching, the latter being the primary driver of outlet mass flux modulation. Flow separation at the control port entrances is shown to throttle the feedback flow and progressively limit oscillation strength at higher inlet flow rates. Restrictive outlet paths induce a differential back pressure that is shown to cause the jet to separate from its attachment wall and bend towards the splitter tip (`secondary separation'). The secondary separation reduces the differential outlet mass flux and introduces a flow curvature that limits the upstream propagation of the back pressure and thus shields the primary jet attachment. The consequence of these effects is that strong oscillations are sustained down to the smallest outlet apertures investigated. The principal contribution is to demonstrate that the assumed coupling between upstream jet attachment and outlet flow split is broken when the outlet aperture is reduced, with significant implications for the design of fluidic oscillators operating with downstream flow impedances.

physics.flu-dyn↗

arXiv2026

Modeling Quantum Geometry for Fractional Chern Insulators with unsupervised learning

Fractional Chern insulators (FCIs) in moire materials present a unique platform for exploring strongly correlated topological phases beyond the paradigm of ideal quantum geometry. While analytical approaches to FCIs and fractional quantum Hall states (FQHS) often rely on idealized Bloch wavefunctions, realistic moire models lack direct tunability of quantum metric and Berry curvature, limiting theoretical and numerical exploration. Here, we introduce an unsupervised machine learning framework to model interacting Hamiltonians directly through the distribution of single-particle form factors. Using a variational autoencoder (VAE), we show that unsupervised learning can not only distinguish FCI and non-FCI states, but also generate new form factors with distinct topological character, not present in the training set. This latent space enables the generation and interpolation of form factors for topological flatbands with Chern number $|C|=1$, enabling the discovery of unobserved many-body states such as charge density waves. Principal component analysis (PCA) further reveals that the dominant patterns in the form factors-reflecting correlations across the Brillouin zone-can be decomposed into components with approximately quantized Chern numbers, providing new insights into the global and topological structure of quantum geometry. Our results highlight the ability of machine learning to generalize and model topological quantum systems, paving the way for the inverse design of form factors with tailored quantum geometry and many-body phases in flatband materials.

cond-mat.str-el↗

arXiv2026

Toward General Digraph Contrastive Learning: A Dual Spatial Perspective

Graph Contrastive Learning (GCL) has emerged as a powerful tool for extracting consistent representations from graphs, independent of labeled information. However, existing methods predominantly focus on undirected graphs, disregarding the pivotal directional information that is fundamental and indispensable in real-world networks (e.g., social networks and recommendations).In this paper, we introduce S2-DiGCL, a novel framework that emphasizes spatial insights from complex and real domain perspectives for directed graph (digraph) contrastive learning. From the complex-domain perspective, S2-DiGCL introduces personalized perturbations into the magnetic Laplacian to adaptively modulate edge phases and directional semantics. From the real-domain perspective, it employs a path-based subgraph augmentation strategy to capture fine-grained local asymmetries and topological dependencies. By jointly leveraging these two complementary spatial views, S2-DiGCL constructs high-quality positive and negative samples, leading to more general and robust digraph contrastive learning. Extensive experiments on 7 real-world digraph datasets demonstrate the superiority of our approach, achieving SOTA performance with 4.41% improvement in node classification and 4.34% in link prediction under both supervised and unsupervised settings.

cs.LG↗

arXiv2026

PeerCheck: Enhancing LLM-Generated Academic Reviews Towards Human-Level Quality

As academic submissions grow, the traditional peer review process struggles to keep up, raising concerns about quality and fairness. A trend of using large language models (LLMs) for assistance has emerged. In this work, we take a critical step toward improving the quality of LLM-generated reviews. We propose the PeerCheck framework, which investigates LLM-human review differences (RQ1) and explores methods to improve LLM-generated review quality (RQ2). We first analyzed the human-written reviews with reviews generated by various LLMs and found that LLMs and humans focus on different terms, e.g., LLMs prioritize theory while humans emphasize methodology and experiments. We further adopt prompt engineering, such as Chain-of-Thought (CoT), and utilize retrieval-augmented generation (RAG) to enhance the LLM-generated reviews towards human-level quality. We find CoT significantly improves the quality of LLM reviews, while we discover an unexpected "RAG paradox," i.e., experiments with RAG produce different results for various LLMs and, in some cases, even reduce review quality. Our comprehensive analysis of LLM-generated academic reviews illustrates both possibilities and limitations, contributing to a more effective, human-aligned review system. Our dataset is available on https://github.com/TrustAIRLab/PeerCheck.

cs.CL↗

arXiv2026

Enhancing Decision-Making with Large Language Models through Multi-Agent Fictitious Play

Large language model (LLM)-based multi-agent systems (MAS) have demonstrated great potential in solving tasks with execution complexity, by distributing subtasks across cooperative agents. However, this divide-and-conquer paradigm falls short on decision-making tasks that are also prevalent in the real world. These tasks require simultaneous reasoning from the stances of all involved stakeholders whose decisions are mutually dependent and thus cannot be solved in isolation. We characterize this challenge as stance entanglement, a form of decision complexity distinct from execution complexity. To address it, we propose Multi-Agent Fictitious Play (MAFP), a novel MAS paradigm that represents stakeholder stances as agents and formulates decision-making as an equilibrium-seeking process. Built on the game-theoretic principle of fictitious play, MAFP iteratively updates each agent's decision by best responding to the empirical mixture of other agents' past decisions. This enables agents to expose and address one another's weaknesses, progressively improving decision quality and robustness. We evaluate MAFP on challenging decision-making tasks that test the capability of deciding strategies for competitive scenarios prior to acting. MAFP outperforms both single-round and multi-round baselines on two complementary metrics, tournament strength and robustness, demonstrating its effectiveness in addressing stance entanglement.

cs.CL↗

arXiv2026

VISUALSKILL: Multimodal Skills for Computer-Use Agents

Computer-use agents (CUAs) approach human-level performance on standardised benchmarks but still struggle on long-horizon tasks and unseen software. Existing skill libraries address this with reusable skills, but represent the skill artifact as text only, despite the visual nature of GUI interaction. We propose VISUALSKILL: a hierarchical multimodal skill, tailored to each target application and organised as a central index over per-topic files, which the agent consumes through a load_topic MCP tool that fetches the relevant topic's text and figures on demand. We construct each skill with a two-stage pipeline that combines authored documentation with live-application UI exploration. On two CUA benchmarks, CUA-World and OSExpert-Eval, a Claude Code CLI agent backed by Claude Opus 4.6 reaches an average score of 0.456 with VISUALSKILL, a +15.3 point absolute lift over the no-skill baseline (0.303). Against a matched text-only skill that is generated from the same source content and differs from VISUALSKILL only in modality, VISUALSKILL yields a further +8.3 point absolute gain over the matched text-only skill (0.373 vs. 0.456), providing direct evidence that retaining visual figures in the skill artifact, rather than verbalizing them away, helps the agent both identify UI elements and verify workflow state after each action. Our code is available at https://github.com/XMHZZ2018/VisualSkills.

cs.CL↗

arXiv2026

Psyzkaller: Learning from Historical and On-the-Fly Execution Data for Smarter Seed Generation in OS kernel Fuzzing

OS Kernel fuzzers such as Syzkaller often struggle to generate syscall sequences that respect intrinsic Syscall Dependency Relations (SDRs), resulting in seeds that either violate kernel constraints or fail to reach deep execution paths. We propose leveraging an N-gram model to learn SDRs from both kernel execution history and ongoing fuzzing results. This enables the fuzzer to capture dependencies in similar kernel versions while adapting to target-specific behaviors, thereby improving the validity of generated seeds. Additionally, we introduce a bidirectional Random Walk strategy to enhance the diversity of generated seeds. We implement this approach in a prototype, Psyzkaller, on top of Syzkaller. Experiments show that, trained with the large-scale DongTing dataset and continuously updated with ongoing fuzzing results, Psyzkaller improves Syzkaller's code coverage by 4.6%-7.0%, triggers 110.4%-187.2% more crashes, and discovers eight previously unknown kernel vulnerabilities. Furthermore, Psyzkaller outperforms state-of-the-art fuzzers such as ACTOR and SyzDescribe in both coverage and crashes.

cs.CR↗

arXiv2026

Kohn--Luttinger Superconductivity in Flat Chern Bands

Recent observations of superconductivity near correlated topological phases in flat bands suggest a facile link between flat-band geometry and electron pairing. In this work, we reveal a geometry-driven Kohn--Luttinger mechanism in which Landau-level-like form factors align the attractive lobe of the RPA-screened Coulomb interaction with the form-factor peak, generating an anomalously strong attractive channel near local band extrema. Using the Skyrmion lattice model as a minimal realization, we show that for spin-unpolarized pairing the form-factor magnitude enforces an emergent momentum-space translational symmetry and selects an extended-$s$ instability concentrated at small Fermi pockets, while for spin-polarized pairing the form-factor phase drives chiral $p$- and $f$-wave order without invoking spin fluctuations. The band-extrema enhancement persists in higher Landau-level analogs and survives finite-temperature screening and Berezinskii--Kosterlitz--Thouless phase fluctuations. Our work establishes quantum geometry as a key organizing principle for unconventional pairing in flat Chern bands.

cond-mat.mes-hall↗

arXiv2026

Tunable Superconductivity in 1313-La$_3$Ni$_2$O$_7$: Suppressed under Compression and Possible $s^{\pm}$ Pairing under Tension

Motivated by recent progress in the 1313-La$_3$Ni$_2$O$_7$ nickelate thin films (Nie et al., Nature {\bf 652}, 628 (2026)), we systematically investigate the effects of both compressive and tensile strain in this system. Self-doping effects between the single-layer (SL) and trilayer (TL) blocks are observed in our studies in both cases, but are most pronounced under tensile strain. We find that superconductivity is unlikely to emerge in the 1313 film on the LSAO substrate even with further hole doping. Remarkably, within the random-phase approximation, under {\it tensile} strain, a robust $s^{\pm}$-wave pairing state emerges in the TL subsystem with sign changes between the small electron-like $σ$ pocket at the $Γ$ point and the small hole-like $γ$ pocket at the M point. These pockets are connected by a wave vector close to $(π,π)$. Our calculations also suggest that superconductivity in 1313-LNO requires an optimally sized $γ$ pocket, because an oversized $γ$ pocket suppresses pairing formation. Overall, our results predict strain-dependent electronic reconstruction in 1313-La$_3$Ni$_2$O$_7$ and provide guiding principles for engineering superconductivity under ambient pressure conditions.

cond-mat.supr-con↗

arXiv2026

Is ChatGPT Fair for Recommendation? Evaluating Fairness in Large Language Model Recommendation

The remarkable achievements of Large Language Models (LLMs) have led to the emergence of a novel recommendation paradigm -- Recommendation via LLM (RecLLM). Nevertheless, it is important to note that LLMs may contain social prejudices, and therefore, the fairness of recommendations made by RecLLM requires further investigation. To avoid the potential risks of RecLLM, it is imperative to evaluate the fairness of RecLLM with respect to various sensitive attributes on the user side. Due to the differences between the RecLLM paradigm and the traditional recommendation paradigm, it is problematic to directly use the fairness benchmark of traditional recommendation. To address the dilemma, we propose a novel benchmark called Fairness of Recommendation via LLM (FaiRLLM). This benchmark comprises carefully crafted metrics and a dataset that accounts for eight sensitive attributes1 in two recommendation scenarios: music and movies. By utilizing our FaiRLLM benchmark, we conducted an evaluation of ChatGPT and discovered that it still exhibits unfairness to some sensitive attributes when generating recommendations. Our code and dataset can be found at https://github.com/jizhi-zhang/FaiRLLM.

cs.IR↗