arXiv subjects

Ying Zhang

Publications and source records attributed to Ying Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Silent and Costly: Characterizing and Detecting Prompt-Caching Mis-Integration in LLM Agents

Large language model (LLM) services process massive volumes of user requests, resulting in substantial inference costs. To reduce both cost and latency, providers have introduced prompt caching. However, correctly integrating prompt caching is non-trivial due to heterogeneous provider-specific APIs and continuously evolving caching semantics. To understand prompt-caching mis-integration in downstream applications, we conduct the first empirical study of 120 prompt-caching-related pull requests from 10 popular open-source agent systems. Our findings show that prompt-caching mis-integrations arise in both request-side misconfiguration and read-side usage miscount. More importantly, most mis-integrations fail silently without causing observable runtime failures. Motivated by these findings, we design CACHELINT, which models a prompt-caching mis-integration as an invocation site together with the preconditions required for correct cache integration. CACHELINT locates caching invocation sites with CodeQL and discharges each precondition either against a dated provider capability oracle or with an LLM guard check. CACHELINT achieves 0.71 precision and 0.78 recall in identifying the 120 labeled mis-integrations across the 10 studied agent systems. Our evaluation further demonstrates the effectiveness of CACHELINT in real-world settings, identifying 32 mis-integrations across 571 agent projects with 0.86 precision. These findings have received 21 developer confirmations and led to 18 upstream fixes, including one in Anthropic's official SDK.

cs.SE↗

arXiv2026

ETAMem: Decoupling and Aligning Heterogeneous Memory for Long-Term Conversational Agents

Long-term memory enables Large Language Models (LLMs) to maintain coherent and personalized interactions across extended conversations. However, existing memory systems face two key challenges: (1) Information Entanglement, where heterogeneous information is mixed within coarse-grained memory units; and (2) Organizational Mismatch, where memory structures based primarily on semantic similarity may overlook other associations. To address these challenges, we propose ETAMem, a multi-view memory framework that decouples heterogeneous conversational information and aligns memory representations and organization with its distinct characteristics. Specifically, heterogeneous information parsing extracts atomic memory units and identifies affective information. View-specific memory structuring then organizes these units into three complementary hierarchical views: Event, Affect, and Time. ETAMem incrementally maintains memory structures with adaptively calibrated thresholds. For retrieval, entropy-based retrieval allocates query-specific budgets across views and performs hierarchical search within each view. Experiments on LoCoMo and LongMemEval-S demonstrate that ETAMem outperforms state-of-the-art baselines, achieving relative accuracy improvements of 12.46% and 7.96%, respectively, while maintaining competitive token efficiency. Further analyses demonstrate the complementary benefits of different memory views.

cs.CL↗

arXiv2026

Photoinduced Odd- and Mixed-Parity Magnetism in Collinear Antiferromagnets

Altermagnetism, characterized by compensated collinear magnetic moments and even-parity momentum-dependent spin splitting, has emerged as a distinct form of magnetic order. However, realizing odd- and mixed-parity magnetism in fully compensated collinear magnets remains challenging. Here, we demonstrate light-induced odd- and mixed-parity magnetism in conventional collinear antiferromagnets. Combining Floquet theory with first-principles calculations, we show that circularly polarized light (CPL) induces odd-parity $f$-wave spin splitting in monolayer CoS and CoSe, whereas elliptically polarized light (EPL) induces $p$-wave spin splitting. Uniaxial strain combined with CPL offers an alternative route to the $p$-wave state. Furthermore, combining an electric field with CPL or EPL generates mixed-parity spin splitting. Light-induced breaking of $\mathcal{PT}$ symmetry, together with spin-orbit coupling, gives rise to a finite anomalous Hall response. Our findings establish a versatile approach to engineering spin splitting and controlling electronic and magnetic properties in compensated antiferromagnets.

cond-mat.mtrl-sci↗

arXiv2026

DeltaTrace: Efficient Signed Attribution for Reasoning Language Models

We study how words in the context support or oppose a language model's response and present DeltaTrace, a novel attribution framework that explains the effect of input tokens on a fixed model response. DeltaTrace measures how the response's log-probability changes when the context tokens being explained are masked. To distribute this difference among source tokens, it constructs operator-specific local rules for finite changes and composes them backward through the computational graph using effective secant slopes. Each token is assigned a signed attribution as the inner product of its embedding displacement and the corresponding backward-propagated slope. Local conservation makes these contributions sum to the measured score difference and preserves their meaning when aggregated over sentences or passages. DeltaTrace supports both attention and linear-attention models, with auxiliary memory growing linearly with sequence length at fixed model dimensions and chunk size. On the evaluated attention model, DeltaTrace achieves the lowest deletion-faithfulness scores (RISE and MAS) on most retrieval and reasoning tasks and the highest needle-in-a-haystack Recall, exceeding FlashTrace by 5.67-24.33 percentage points. On the evaluated hybrid attention and linear-attention model, DeltaTrace lowers RISE on 9 tasks and MAS on 12 relative to FlashTrace.

cs.LG↗

arXiv2026

Multi-Perspective Evidence Synthesis and Reasoning for Unsupervised Multimodal Entity Linking

Multimodal Entity Linking (MEL) is a fundamental task in data management that maps ambiguous mentions with diverse modalities to the multimodal entities in a knowledge base. However, most existing MEL approaches primarily focus on optimizing instance-centric features and evidence, leaving broader forms of evidence and their intricate interdependencies insufficiently explored. Motivated by the observation that human expert decision-making process relies on multi-perspective judgment, in this work, we propose MSR-MEL, a Multi-perspective Evidence Synthesis and Reasoning framework with Large Language Models (LLMs) for unsupervised MEL. Specifically, we adopt a two-stage framework: (1) Offline Multi-Perspective Evidence Synthesis constructs a comprehensive set of evidence. This includes instance-centric evidence capturing the instance-centric multimodal information of mentions and entities, group-level evidence that aggregates neighborhood information, lexical evidence based on string overlap ratio, and statistical evidence based on simple summary statistics. A core contribution of our framework is the synthesis of group-level evidence, which effectively aggregates vital neighborhood information by graph. We first construct LLM-enhanced contextualized graphs. Subsequently, different modalities are jointly aligned through an asymmetric teacher-student graph neural network. (2) Online Multi-Perspective Evidence Reasoning leverages the power of LLM as a reasoning module to analyze the correlation and semantics of the multi-perspective evidence to induce an effective ranking strategy for accurate entity linking without supervision. Extensive experiments on widely used MEL benchmarks demonstrate that MSR-MEL consistently outperforms state-of-the-art unsupervised methods. The source code of this paper was available at: https://anonymous.4open.science/r/MSR-MEL-C21E/.

cs.CL↗

arXiv2026

Ensemble-based Deep Multilayer Community Search

Multilayer graphs, consisting of multiple interconnected layers, are widely used to model diverse relationships in the real world. A community is a cohesive subgraph that offers valuable insights for analyzing (multilayer) graphs. Recently, there has been an emerging trend focused on searching query-driven communities within the multilayer graphs. However, existing methods for multilayer community search are either 1) rule-based, which suffer from structure inflexibility; or 2) learning-based, which rely on labeled data or fail to capture layer-specific characteristics. To address these, we propose EnMCS, an Ensemble-based unsupervised (i.e., label-free) Multilayer Community Search framework. EnMCS contains two key components, i.e., HoloSearch which identifies potential communities in each layer while integrating both layer-shared and layer-specific information, and EMerge which is an Expectation-Maximization (EM)-based method that synthesizes the potential communities from each layer into a consensus community. Specifically, HoloSearch first employs a graph-diffusion-based model that integrates three label-free loss functions to learn layer-specific and layer-shared representations for each node. Communities in each layer are then identified based on nodes that exhibit high similarity in layer-shared representations while demonstrating low similarity in layer-specific representations w.r.t. the query nodes. To account for the varying layer-specific characteristics of each layer when merging communities, EMerge models the error rates of layers and true community as latent variables. It then employs the EM algorithm to simultaneously minimize the error rates of layers and predict the final consensus community through iterative maximum likelihood estimation. Experiments over 10 real-world datasets highlight the superiority of EnMCS in terms of both efficiency and effectiveness.

cs.SI↗

arXiv2026

Label-Centric Knowledge Retrieval for Zero-Shot Text Classification

Flagship large language models (LLMs) offer rich general knowledge and strong reasoning capabilities, but their high inference cost and latency limit their use for large-scale data labeling. We study zero-shot text classification with semantically meaningful labels and task descriptions, without task-specific labeled examples or classifier feedback. We propose \textbf{SLKR} (Semantic Label Knowledge Retrieval), a label-centric retrieval-augmented generation framework that separates offline knowledge construction from lightweight online inference. Offline, a flagship LLM independently generates label definitions, retrieval keywords, and boundary rules for potentially confusable label pairs, forming a reusable label-knowledge store. Because this construction operates over the label space rather than the input corpus, its one-time flagship cost can be amortized across large-scale inference. Online, hybrid semantic and lexical retrieval first identifies the top-$K$ candidate labels and then selects up to $M$ input-relevant boundary rules connecting candidate-label pairs. A lightweight LLM performs the final classification in a single call using the retrieved definitions and rules. Across three public benchmarks and a real-world multilingual book-classification dataset, SLKR achieves the best results on all four datasets with Qwen3.5-4B and on three of four with Ministral-3-3B, while using compact input contexts and no task-specific labeled examples. Boundary ablations further show that retrieving a small set of input-relevant rules preserves performance close to using all boundaries with substantially less context, supporting efficient reuse of flagship-generated label knowledge for lightweight classification.

cs.AI↗

arXiv2026

VeriNC: Finding Design Risks of In-Network Computing Systems

The emergence of programmable switches has brought in-network computing (INC) into the spotlight in recent years. By offloading computation directly onto the data transmission process, INC improves network utilization, reduces latency to sub-RTT levels, saves link bandwidth, and maintains throughput. However, INC disrupts the transparency of traditional networks, forcing developers to consider network exceptions like packet loss and out-of-order. If not properly handled, these exceptions can lead to violations of application properties, such as cache consistency and lock exclusion. Usual testing cannot exhaustively cover these exceptions, raising doubts about the correctness of INC systems and hindering their deployment in the industry. This paper presents VeriNC, the first general-purpose tool for verifying INC systems. VeriNC provides a high-level specification language and saves developers 67.2% lines of code on average. To help better understand the behavior of the system, VeriNC offers configurable network environments. VeriNC enables developers to express INC-specific correctness properties. VeriNC translates developer-specified systems into state transition representations, performs model checking to detect potential design risks, and reports violation traces to developers. We propose optimizations for INC-specific scenarios to address the challenge of state space explosion. We modeled INC systems across four application domains and identified design risks with VeriNC in seconds. VeriNC has also been adopted to guide the design of a new INC protocol. Based on our verification experience, we summarize lessons that help develop a correct INC protocol. We further reproduce them in real systems to confirm the validity of our verification result.

cs.DC↗

arXiv2026

Correspondences as Decisions: JevNexus for Decision-Centric Schema Matching

Schema matching increasingly uses generative language models to rerank retrieved column candidates, although the underlying task is a bounded correspondence decision. We present JevNexus, which combines typed pairwise decisions with schema/instance evidence and invokes listwise refinement only when the evidence disagrees and the fused margin is small. The evaluation covers 561 cases from six benchmark families. JevNexus obtains dataset-macro MRR and Hits@1 of 0.930 and 0.909, compared with 0.926 and 0.903 for Magneto, while reducing mean latency from 123.452 to 15.929 seconds (7.750). Paired analysis finds no statistically significant difference in either MRR or Hits@1. The gate invokes listwise refinement for only 5.665% of source columns and avoids the degradation caused by unconditional refinement. Code and experimental artifacts are available at https://github.com/RazeenLI/JevNexus.

cs.DB↗

arXiv2026

Anomalous Hall enhancement in fully compensated non-collinear antiferromagnetic Cr-doped Mn3Sn

The non-collinear antiferromagnet (AFM), Mn3Sn, exhibits ultrafast spin dynamics and negligible stray fields, making it a promising candidate for fast and energy-efficient spintronic devices. However, its practical applications remain limited by the relatively small Hall response compared with ferromagnets. Here, we demonstrate through both first-principles calculations and experiments that Cr-doping is a promising strategy to enhance the anomalous Hall transport in Mn3Sn without breaking the antiferromagnetic order. Our first-principles calculations reveal an effective Cr-doping window that enhances the anomalous Hall conductivity while largely preserving the host coplanar inverse triangular AFM order. Berry curvature and Wannier Hamiltonian analyses further reveal that Cr substitution amplifies the pre-existing Berry curvature hotspots through coordinated changes in the energy alignment, local interband gap, and Mn-derived orbital weight of the relevant bands. Experimentally, we have found that Cr doping not only supports our computational predictions (enhancement of the anomalous Hall conductivity), but also suppresses the transition to helical or spin-glass phases at low temperatures. These findings demonstrate that Cr-doping is an effective strategy for stabilizing the AFM order while simultaneously improving the anomalous Hall transport properties over a wide temperature range in Mn3Sn.

cond-mat.mtrl-sci↗

arXiv2026

Enhancing LLMs with Cognitive-Affective Personality Inference for Simulating Human Social-Psychological Behavior

Large language models are increasingly used to simulate human participants in social and behavioral studies, yet static persona prompting typically maps a participant profile and an experimental scenario directly to a response, entangling stable dispositions with situation-specific interpretations. To address this limitation, we introduce \textbf{SPIN}, a cognitive-affective personality system-inspired inference pipeline for simulating human social-psychological behavior. Specifically, SPIN implements this structured inference process through three zero-shot LLM calls that compile a task-blind participant core, elicit condition-specific cognitive-affective states, and read out decisions from those states, thereby reusing stable personality structure while routing each trial-specific response through an explicit state representation. We evaluate SPIN on two reconstructed social-psychological study families spanning uncertainty reasoning and pluralistic ignorance, across four base LLMs. Compared with blank, demographic, narrative, and chain-of-thought prompt variants, SPIN consistently delivers the strongest overall alignment performance across base LLMs and study families. Ablations and state analyses further show that both personality compilation and structured state elicitation contribute to the gains, and that the elicited states shift interpretably across informational and normative conditions. These results suggest that structured personality-state inference can improve benchmark-level behavioral alignment beyond richer persona descriptions or generic multi-step reasoning.

cs.SI↗

arXiv2026

Nuclear mass table in deformed relativistic Hartree-Bogoliubov theory in continuum, III: nuclei with $8 \leq Z \leq 120$

The mass table in the deformed relativistic Hartree-Bogoliubov theory in continuum (DRHBc) with the PC-PK1 density functional has been established for nuclei with $8 \leq Z \leq 120$, extended from the previous works for even-even nuclei [Zhang et al. (DRHBc mass table collaboration), At. Data Nucl. Data Tables 144, 101488 (2022)] and for even-$Z$ nuclei [Guo et al. (DRHBc mass table collaboration), At. Data Nucl. Data Tables 158, 101661 (2024)]. The calculated binding energies, two- and one-nucleon separation energies, root-mean-square (rms) radii of neutron, proton, matter, and charge distributions, quadrupole deformations, neutron and proton Fermi surfaces, and the blocked neutron (proton) orbitals of odd-$N$ ($Z$) nuclei are tabulated and compared with the available experimental data. A total of 9495 nuclei are predicted to be bound, with an rms deviation of 1.444 MeV from the 2380 mass data. Good agreement with the available experimental pairing gaps, $α$ decay energies, and charge radii is also achieved. The accuracies of the calculated nuclear masses and nucleon separation energies as well as the prediction for drip lines are compared with those obtained by other relativistic and nonrelativistic density functional calculations. It turns out that the DRHBc theory with PC-PK1 provides one of the best microscopic descriptions for nuclear masses. The systematics of nucleon separation energies, pairing gaps, pairing energies, two-nucleon gaps, $α$ decay energies, rms radii, quadrupole deformations, potential energy curves, neutron density distributions, and neutron mean-field potentials are discussed.

nucl-th↗

arXiv2026

DramaChain Bench: An End-to-End Benchmark for Short-Drama Generation

Commercial short-drama production follows a multi-stage chain: script, storyboard, keyframe imagery, shot-level video, and the finished short drama. Most existing benchmarks evaluate solely the video-generation stage using pre-authored inputs instead of real upstream pipeline outputs. This leaves two critical questions unanswerable: whether each stage adheres to the original script intent (rather than only its immediate input prompt), and whether disparate shots remain coherent after assembly into multi-episode releases. We present DramaChain Bench, the first short-drama benchmark that evaluates every stage of the complete production chain. It is built upon three in-house systems sharing one dimension system, DramaChain Dimensions: five evaluation axes instantiated at every stage, resolving into 63 leaf dimensions. DramaChain Agent is calibrated against commercial short-drama platforms in both workflow and finished short-drama quality, enabling stage-wise fair comparison across models. DramaChain Labeling System has each of the 5,785 items scored independently by three professional annotators, with all defects spatio-temporally localised and selected from a predefined defect list. This process produces 17,488 valid scores and 255,925 traceable attribution records. The human annotations confirm that upstream defects cascade across the pipeline, demonstrating that final episode quality is not governed by video generation alone. DramaChain Agentic Judge then scores every leaf dimension automatically, gathering evidence over multiple agentic rounds before judging against a per-item checklist; it reproduces the model ranking at a mean PLCC of 0.918, enough to admit new models at no annotation cost.

cs.AI↗

arXiv2026

A Proof of Fraenkel's Conjecture

Fraenkel's conjecture asserts that a partition of the integers into at least three Beatty sequences with distinct moduli has the binary densities $1,2,4,\ldots,2^{m-1}$, normalized by $2^m-1$. We prove the conjecture through a dimension-free intermediate statement: every such partition contains a component of density at least 1/3. After reducing the partition to primitive common-period data, Fourier cancellation produces a finite inverse-sine system. We prove that no such system can exist when every density is below 1/3. The proof combines a divisor-concentration identity with uniform analytic estimates and three exact finite verifications, all carried out with integer or rational arithmetic. The component supplied by the density bound has mean spacing at most three. Deleting it preserves balance, and every surviving periodic balanced set is again a rational Beatty set. Induction determines the surviving binary scales, while a two-sequence disjointness criterion forces the deleted density to be the next binary scale. This yields the asserted density pattern.An eventual Beatty representation of balanced binary indicators extends the classification to one-sided balanced sequences on at least three letters with distinct positive densities, proving the balanced-sequence conjecture of Altman, Gaujal, and Hordijk.

math.NT↗

arXiv2026

MedVL-SAM2: A unified 3D medical vision-language model for multimodal reasoning and prompt-driven segmentation

Recent progress in medical vision-language models (VLMs) has achieved strong performance on image-level text-centric tasks such as report generation and visual question answering (VQA). However, achieving fine-grained visual grounding and volumetric spatial reasoning in 3D medical VLMs remains challenging, particularly when aiming to unify these capabilities within a single, generalizable framework. To address this challenge, we proposed MedVL-SAM2, a unified 3D medical multimodal model that concurrently supports report generation, VQA, and multi-paradigm segmentation, including semantic, referring, and interactive segmentation. MedVL-SAM2 integrates image-level reasoning and pixel-level perception through a cohesive architecture tailored for 3D medical imaging, and incorporates a SAM2-based volumetric segmentation module to enable precise multi-granular spatial reasoning. The model is trained in a multi-stage pipeline: it is first pre-trained on a large-scale corpus of 3D CT image-text pairs to align volumetric visual features with radiology-language embeddings. It is then jointly optimized with both language-understanding and segmentation objectives using a comprehensive 3D CT segmentation dataset. This joint training enables flexible interaction via language, point, or box prompts, thereby unifying high-level visual reasoning with spatially precise localization. Our unified architecture delivers state-of-the-art performance across report generation, VQA, and multiple 3D segmentation tasks. Extensive analyses further show that the model provides reliable 3D visual grounding, controllable interactive segmentation, and robust cross-modal reasoning, demonstrating that high-level semantic reasoning and precise 3D localization can be jointly achieved within a unified 3D medical VLM.

cs.CV↗

arXiv2026

OceanMind: A multi-agent AI system for ocean diagnosis

Time-dependent, three-dimensional (3D) oceanic multi-variables define coherent states of the evolving ocean to facilitate ocean diagnosis and advance ocean science to better inform environmental and hazard management. However, extracting quantitative evidence from these variables requires substantial and complex analytical effort. We introduce OceanMind, a multi-agent AI system that directly couples large language models (LLMs) with comprehensive time-dependent 3D ocean states for swift and effective diagnosis. OceanMind organizes the analytical process into four coordinated complexity stages: Query Routing, Skill-based Planning, Tool Execution, and Evidence-based Summary Generation. Specialized agents interpret user requests, construct and execute multi-step computational workflows, and synthesize quantitative evidence. To ensure reliable workflow construction, 63 reusable ocean-specific analysis skills serve as procedural manuals that guide the LLM agent in selecting data, conducting diagnostics, and applying analytical tools. With reflection and replanning mechanisms that use execution feedback to repair invalid plans, OceanMind ensures reliable analysis workflows across diverse needs. On a benchmark of 240 computational-workflow queries spanning the four stages, OceanMind outperformed general ReAct agents with the same registered tool pool, achieving relative improvements of 41.2% in effectiveness and 21.5% in efficiency. Beyond the benchmark, OceanMind reproduced published oceanic diagnostics, validated hypotheses, and supported environmental decision-making over global oceans. Overall, OceanMind advances LLMs by integrating them with time-dependent 3D ocean analysis, enabling scientific interpretation and enhancing formulation of environmental policies based on quantitative ocean evidence.

physics.ao-ph↗

arXiv2026

A Joint Information-Theoretic and Graph Learning Framework for Sensor Placement and Macroscopic Traffic State Reconstruction

Traffic sensor placement in large urban networks must balance traffic monitoring needs with limited installation, maintenance, and data-management budgets. This study develops a framework for preserving network-wide traffic information while reconstructing flow and density at unobserved links and the macroscopic fundamental diagram (MFD). The framework combines normalized mutual information (NMI) with Inductive Graph Neural Network Kriging (IGNNK). NMI quantifies lag-aware dependence between joint flow-density states, while a greedy algorithm selects sensors by balancing weighted information coverage against redundancy. An IGNNK-based bidirectional graph-diffusion model reconstructs unobserved link states and generates the MFD. The framework was evaluated against a principal component analysis (PCA) benchmark using seven hours of dynamic traffic assignment simulation data for the Chicago network, with 4,805 directed links and 420 one-minute intervals split into training, validation, and independent testing subsets. Reconstruction performance for both PCA and NMI-IGNNK varied substantially over the sensor ratios of 0.5% to 30%. PCA generally yielded lower root mean squared error (RMSE) over the 0.5%-3% range, but its errors were nonmonotonic and reached their minimum at 2%. At 3%, normalized joint RMSE was similar for PCA and NMI-IGNNK (0.100 and 0.105, respectively). NMI-IGNNK remained applicable up to 30% coverage and generally improved as coverage increased, reducing testing-subset normalized joint RMSE from 0.212 at 0.5% to 0.043 at 30%, with consistent trends over the complete analysis period. The framework integrates lag-aware, information-theoretic sensor placement based on joint flow-density states with directed graph-based reconstruction, enabling a broad range of sensor budgets in large-scale networks without strict limits on the number of sensors.

eess.SP↗

arXiv2026

One Latent, Many Tokens: Jointly Learning Compressed Embeddings for Efficient Language Diffusion

Most continuous diffusion language models process one latent position per token at each sampling step, making generation expensive. Two-stage methods lower the cost by reducing the latent length, but they fix the compressed embedding space before training the diffusion model. Embeddings from the fixed space can be difficult to model with diffusion and decode reliably into tokens, which limits generation quality after compression. To address this problem, we introduce JPEG-DLM (Joint-embedding Prediction for Efficient Generation with Diffusion Language Model), which jointly trains a compressor, a flow matching model and a decoding module. With joint-embedding prediction, JPEG-DLM learns compressed embeddings that are more structured, easier to model with diffusion and reliably decodable into tokens. JPEG-DLM achieves the lowest mean Gen-PPL and highest throughput among recent diffusion and flow models on LM1B and OWT. At a compression rate of 0.5 on OWT, it reaches a Gen-PPL of 34.52 and approximately 2.3 times ELF's throughput. These results suggest that jointly learning compressed embeddings offers a promising path toward efficient diffusion language modeling. Code will be released soon.

cs.AI↗