arXiv subjects

Hao Sun

Publications and source records attributed to Hao Sun.

At least 19 recordsLinked to original sources

arXiv2026

Discovering hierarchical governing equations for community-structured complex network dynamics with LLM agents

Governing equations provide an interpretable description of the mechanisms underlying complex network dynamics, yet their discovery from observational data remains challenging when shared community-level dynamics coexist with individual-level variations and the relevant functional structures may be absent from the initial candidate library or not explicitly specified in advance. Here we formulate this task as a hierarchical law-discovery problem and introduce SAGE, a large language model-guided self-reflective agent framework for discovering community-shared governing laws and individual-specific dynamical variations. SAGE organizes equation discovery into an iterative Library-Discovery-Review process that first identifies shared community-level structures and then resolves node-specific residual dynamics, while adaptively refining the candidate space to introduce missing dynamical terms when needed. This hierarchical and interpretable formulation enables SAGE to distinguish shared community-level dynamics from individual-level variations and to reveal the mechanisms underlying complex network dynamics. Across two synthetic networks and four real-world networks spanning mobile communication, brain electrophysiology, power grids and urban rail transit, SAGE recovers compact and interpretable governing equations, reproduces observed system dynamics and outperforms representative equation-discovery baselines. The framework remains robust under incomplete observations and can transfer community-level governing laws to nodes with entirely missing observations. These results establish a hierarchical and adaptive approach for disentangling shared and individual dynamical mechanisms in community-structured complex networks.

physics.soc-ph↗

arXiv2026

DriftingPlanner: Drifting Generative Models for One-Step Trajectory Planning in End-to-End Autonomous Driving

End-to-end planners decode the ego trajectory either by coordinate regression, which is fast but blends plausible maneuvers into one trajectory that may follow none of them, or by diffusion and flow matching, which model the trajectory distribution but need several inference steps. Drifting models move this iteration into training: a one-step generator learns to follow an attraction--repulsion field between generated and ground-truth samples. Drifting was developed for class-conditional image generation, where every class supplies many positive samples. In planning, every scene is its own condition. Pooling scenes would attract the generated trajectories of one scene to the ground truth of unrelated scenes, whereas computing the field per scene leaves a single positive, the expert trajectory. Mutual repulsion then holds the generated trajectories on a shell around it, so any single generated trajectory deviates from the ground truth. We present DriftingPlanner, a trajectory planner trained with the drifting loss alone, without any coordinate regression term. It computes the field per scene and removes the shell at inference: it decodes every noise vector together with its negation in one batched pass, and averaging each antithetic pair cancels the linear part of the displacement from the shell center. On NAVSIM navtest, DriftingPlanner reaches a state-of-the-art 93.75 PDMS, and with single-step inference its drifting decoder comes within 0.13 PDMS of the best diffusion or flow matching setting, which needs ten inference steps.

cs.CV↗

arXiv2026

Arbitrary-order exceptional points in a nanomechanical cavity

Exceptional points (EPs), non-Hermitian spectral singularities, shape the evolution dynamics ranging from topological mode control to high-precision sensing. Higher-order EPs unlock enriched and sharpened spectral topology beyond the minimal second-order counterpart, thus expanding capabilities in those domains. However, the intrinsic topological fragility of higher-order EPs has long hindered their robust experimental realization. Here, we present a scalable architecture that implements arbitrary-order EPs via an optically mediated recurrent network comprising a single nanomechanical resonator and an ensemble of digital synthetic resonators. We experimentally validate this architecture's scalability for both linear and nonlinear mechanical EPs, demonstrating EPs up to the seventh order. Moreover, we reveal that the fundamental noise component, in contrast to some technical-noise contributions, shares the same system coupling channel and undergoes identical root-response amplification as the measured signal. The co-amplification mechanism sets the fundamental sensitivity limit near EPs of any order, as verified by direct signal-to-noise ratio measurements. Our work establishes a general platform for exploring higher-order EP-based phenomena while clarifying the fundamental boundary of non-Hermitian sensitivity enhancement. More broadly, it facilitates a connection between non-Hermitian physics and AI-inspired recurrent architectures across diverse physical systems.

physics.optics↗

arXiv2026

HABIT-Bench: Evaluating Habit Induction from Longitudinal Weak Evidence in Agent Memory

In long-term interactions with large language model (LLM) agents, users reveal habits through their choices and actions. These habits capture implicit user information that agents must infer across conversations. However, individual observations may provide only partial evidence about a habit or when it applies. Retaining past interactions alone does not ensure that agents can infer habits and apply them appropriately. We introduce HABIT-Bench to evaluate how well agent memory systems infer habits from weak evidence and apply them to later requests. We construct synthetic user histories from task-oriented dialogue data and controlled habit graphs. The histories span hundreds of sessions across food, finance, software, and travel. Four-choice tasks assess agents' ability to infer user habits and apply them appropriately. They include temporary exceptions, changes in habits, and suggestions that users have not accepted. Sampled tasks undergo independent human review. We evaluate eight memory methods with Qwen3-235B-A22B and GLM-5.3-Flash. The best-performing memory method achieves 33.61\% macro accuracy across the four domains. Performance remains limited even when annotated evidence is supplied directly. These results expose a persistent gap between access to past interactions and the ability to infer and appropriately use user habits.

cs.AI↗

arXiv2026

From Evidence to Effect: Authority Semantics and Runtime Infrastructure for Stateful Agents

Stateful agents reuse artifacts after producing executions and permissions change. We formalize authority-sufficient observations and durable effects bound to execution and material identities. WTB implements this interface through runtime adapters, shared evidence, and transactional publication/recovery. Six study families separate the mechanism from its integration. Raw and typed evidence both solve 32/32 authority cases, with model-dependent planning effects. Fixed-intent enforcement blocks six unsafe proposals and executes 12 eligible authorized intents. Complete controls match WTB's capability. Paid integration yields 176/210 accepted benchmark-source stages, including 19/30 publication stages, recovery on 8/8 primary SWE repositories, and the most complete continuous trajectories on each of three source tasks. The findings connect authority information, effect admission, and infrastructure reuse in stateful agents.

cs.MA↗

arXiv2026

Deep Fusion Memory: A Chunk-Level Parametric Memory for Large Language Models

Large language models need effective ways to acquire and use external knowledge beyond what they retain from pretraining. Continued pretraining risks catastrophic forgetting, while retrieval at inference time can add substantial overhead. External parametric memory avoids online retrieval, but constructing its training data requires extensive offline search, and output-level interpolation provides only shallow fusion. We introduce Deep Fusion Memory (DFM), a parametric memory architecture that reduces this construction workload and incorporates learned memory into intermediate hidden states. A causal memory predictor learns to generate memory from the preceding context, using retrieved chunk embeddings from the corpus as supervision. The predicted memory is shared across tokens within each chunk and fused into their hidden states through gated cross-attention. Memory prediction and fusion are trained end to end with the backbone frozen, and inference requires no explicit retrieval. Compared with existing parametric memory approaches, which often construct token-level memory targets, DFM requires $32\times$ fewer retrieval queries and $2048\times$ fewer retrieved items to construct its training data. Across six QA tasks, DFM performs better over the strongest baseline by 12.71% on Mistral-7B-v0.3 and 11.19% on Qwen3-8B-Base. Beyond QA, domain adaptation yields average gains of 6.51 and 4.38 points over Qwen3-4B-Base and Qwen3-8B-Base, respectively.

cs.CL↗

arXiv2026

VTrajEmbed: Aligning Text with Video Object Trajectories via Spatiotemporal Embeddings

Fine-grained alignment between text and video object trajectories is a core challenge in multimodal understanding. Existing embedding models excel at global retrieval, yet their single global representation entangles multiple objects and their spatiotemporal dynamics, limiting direct object-level grounding. To address this issue, we propose VTrajEmbed, an embedding model that decomposes a video into object trajectories and aligns each trajectory embedding with text for object-level grounding through retrieval. Our main contributions: (1)Slow-Fast Trajectory Encoding, a dual-branch architecture that establishes trajectory representations by densely sampling low-resolution frames for high-frequency motion cues and sparsely selecting high-resolution frames for scene context, aggregated by a shared LLM backbone; (2)Mask-Guided Contrastive Learning, a training objective built upon these representations that couples an intra-group loss and a batch-level bidirectional loss under a relation mask, excluding unverified relations and preventing confirmed matches from becoming false negatives; (3)Text2Traj-Bench, a trajectory-level text-video benchmark with model-assisted construction and human-verified query--trajectory relations. Experimental results show that VTrajEmbed-5B improves text-to-trajectory and trajectory-to-text R@1 over Qwen3-VL-Embedding-8B by 15.48 and 18.05 percentage points, respectively, and achieves 49.48% average R@1 across five global retrieval benchmarks, compared with 42.86% for Qwen3-VL-Embedding-8B.

cs.CV↗

arXiv2026

ReferFlow: Beyond Final-Target Localization for Complex Referring Expressions

Multimodal Large Language Models have made significant progress in referring expression comprehension (REC), yet most REC benchmarks still focus on final-target localization. Complex queries often involve intermediate visual references with different dependencies on prior grounding, but final-only evaluation leaves this process unexamined. To address this gap, we formulate trajectory-level REC, in which models receive an ordered sequence of grounding steps constructed from a single complex query and localize the target regions associated with each step. We instantiate this setting in ReferFlow-Bench, which characterizes structured region transitions through four trajectory types, namely Refinement, Shift, Reset, and Mixture. Experiments show that current models struggle to consistently ground intermediate visual references across successive steps. We further propose ReferFlow, which decouples interleaved text-latent trajectory generation from coordinate decoding. Specifically, Stateful Region Latent Reasoning represents each target region as an object-indexed state composed of continuous latents, with explicit visual and spatial supervision to preserve region information for subsequent grounding. Trajectory-Aware Parallel Decoding then recovers all coordinate blocks in parallel through four geometry-constrained refinement rounds, with each block conditioned on its corresponding trajectory prefix, thereby reducing within-box sequential dependencies and eliminating cross-box autoregression. We further apply Grounding-Aware GRPO to promote consistent grounding across steps and learn adaptive trajectory generation from complex REC queries. ReferFlow achieves leading performance on general REC benchmarks and ReferFlow-Bench.

cs.CV↗

arXiv2026

ROI-Reasoning: Rational Optimization for Inference via Pre-Computation Meta-Cognition

Large language models (LLMs) can achieve strong reasoning performance with sufficient computation, but they do not inherently know how much computation a task requires. We study budgeted inference-time reasoning for multiple problems under a strict global token constraint and formulate it as an Ordered Stochastic Multiple-Choice Knapsack Problem (OS-MCKP). This perspective highlights a meta-cognitive requirement---anticipating problem difficulty and allocating computation strategically. We propose ROI-Reasoning, a two-stage framework that endows LLMs with intrinsic, budget-aware rationality. First, Meta-Cognitive Fine-Tuning teaches models to predict reasoning cost levels and decide whether a problem is worth attempting before generation. Next, Rationality-Aware Reinforcement Learning optimizes sequential decision making under a hard token budget, allowing models to learn long-horizon allocation strategies. Experiments across mathematical reasoning, multiple-choice question answering, and code generation show that ROI-Reasoning improves scores and generally reduces reference-order regret under tight budgets, with up to a \textbf{417\%} relative score improvement over the strongest same-backbone baseline in mathematics.

cs.AI↗

arXiv2026

Large language models for partial differential equation workflows

Numerical methods and deep learning have advanced the modelling and simulation of systems governed by partial differential equations (PDEs), while formulating equations, configuring solvers, and translating simulations into design and control decisions continue to demand substantial domain expertise and computational resources. Large language models (LLMs) offer opportunities to automate and coordinate these tasks by combining scientific knowledge, mathematical reasoning, code generation, and tool use. Here we review LLM-assisted PDE research across three stages, including discovery, solving, and optimization. We examine how LLMs interact with physical knowledge, numerical methods, and computational feedback to formulate models, implement solution procedures, and guide design and control. We also review existing benchmarks and evaluation practices for assessing scientific validity, computational performance, and workflow reliability. Finally, we discuss three interconnected challenges and opportunities: developing mathematical and physical reasoning tailored to specific PDE problems; building transferable domain expertise through domain-specific and multimodal post-training; and integrating active experimental design with scientific validation for open-ended problems. These directions outline a path towards a more substantive role for LLMs in developing and testing scientific ideas and computational methods across PDE research.

cs.AI↗

arXiv2026

Thermal contact and direct detection in a broken $Z_4$ Majorana dark matter model

We consider a broken $Z_4$ Majorana dark matter model in which the fermion mass is generated by a real singlet scalar. For small scalar mixing, freeze-out is dominated by $χχ\to h_2h_2$, whereas momentum exchange with the Standard Model bath becomes inefficient. In this region, the usual assumption $T_χ=T$ is not always adequate. We therefore evolve the dark matter abundance and kinetic temperature simultaneously. Although $T_χ$ is still close to the bath temperature around chemical freeze-out, the later departure from kinetic equilibrium reduces the residual $p$-wave annihilation rate and leaves a larger relic abundance. In the sub-TeV region considered here, the difference between the two-moment and single-temperature calculations reaches $37.1\%$ at fixed parameters. Refitting the Yukawa coupling to reproduce $Ω_χh^2=0.12$ increases the spin-independent scattering cross section by up to $18.6\%$. A momentum-resolved Fokker--Planck calculation gives a similar shift, $17.7\%$, at a representative point. The LZ upper limit on the scalar mixing changes by less than $0.3\%$, since kinetic equilibrium is already well maintained in that part of parameter space. The lower mixing boundary obtained from the comparison of the two relic calculations therefore indicates where the common-temperature approximation loses accuracy; it is not an experimental lower bound. These results apply to the specified low-reheating broken-phase cosmology and to the region where threshold and strong nonrelativistic-potential effects remain small.

hep-ph↗

arXiv2026

Solar Capture and Suppressed Annihilation of Inelastic Scalar Dark Matter

Solar capture of inelastic dark matter does not by itself fix the annihilation signal. We consider a scalar low-energy model with a $B-3L_τ$ current, motivated by the high-energy recoil candidate reported by LUX-ZEPLIN. For $m_χ=840~\mathrm{GeV}$, a $320~\mathrm{keV}$ splitting, and a $160~\mathrm{MeV}$ mediator, the public LZ response gives 0.965 expected accepted events, while micrOMEGAs gives $Ωh^2=0.11999834$. Finite-momentum nuclear responses leave iron as an important solar-capture target despite the near cancellation of its coherent charge at zero momentum transfer. The captured population remains extended after thermal nuclear motion, elastic and inelastic scattering, and excited-state decay are included, with $r_{\rm eff}=0.215R_\odot$. Same-state annihilation is open, but the inferred rate is only 0.0404 of the public IceCube $W^+W^-$-calibrated proxy and is smaller than the corresponding isothermal estimate by a factor of 628. For this benchmark, substantial solar capture is accompanied by a weak neutrino signal.

hep-ph↗

arXiv2026

Case-Level Verification in Scanner-LLM Cascades: Overcoming the Alert Aggregation Bottleneck to Expand the FRR-TPR Trade-off Space

Dynamic Application Security Testing (DAST) scanners achieve high recall but also produce a large number of false positives, resulting in substantial manual triage costs. Large Language Models (LLMs), when used for independent detection, achieve extremely high recall (95.4%-100%) but also exhibit prohibitively high false positive rates (49.6%-85.0%), precluding their use as standalone replacements for scanners. A natural solution is a two-stage cascade consisting of scanner detection followed by LLM verification. However, a verification-granularity issue that has long been overlooked in practice creates a structural bottleneck: alert aggregation binds multiple true and false cases into a shared decision unit, such that removing a false positive inevitably eliminates true positives aggregated within the same alert group. This creates a trade-off bottleneck between the False-positive Reduction Rate (FRR) and the True-positive Rate (TPR). We formalize this bottleneck by showing that the alert-level false-positive set is a subset of the case-level false-positive set, and introduce a Case-Level, per-case verification strategy that shifts the decision granularity from the alert level to the instance level, independently replaying HTTP requests and making an independent determination for each detected case. Evaluation on the dual testbeds of Damn Vulnerable Web Application (DVWA) and WebGoat shows that the empirically best Alert-Level operating point achieves FRR=42.86% (TPR=51.7%). The Case-Level Baseline achieves FRR=47.6%, an improvement of 4.7 percentage points (+4.7 pp), while the Case-Focused Evidence Verification Prompt (CEV-Prompt) increases TPR from 55.2% to 62.1% at the same FRR.

cs.CR↗

arXiv2026

Beyond the Flat-Band Paradigm: Open Orbits as a Route to Correlated States in Dispersive Moiré Minibands

The realization of fractional Chern insulators has largely relied on isolated flat bands in complex bilayer or multilayer moiré systems with ideal quantum geometry. Whether such phases can emerge in genuinely dispersive bands, and by what mechanism, remains an open question. Here, we propose a mechanism for magnetic-field-induced strong correlations in the highly dispersive bands of monolayer graphene on hexagonal boron nitride. In the low-field classical-to-quantum crossover regime of the second moiré miniband, we show that the interplay of the moiré potential and weak magnetic fields (B < 1.5 T) profoundly alters electron dynamics. Saddle-point van Hove singularities produce extended open orbits that strongly suppress carrier propagation, potentially favoring a regime in which Coulomb interactions become comparatively more important, while weak magnetic fields lift valley degeneracy and trigonal warping redistributes Berry curvature. High-resolution magnetotransport and temperature-dependent measurements reveal fractional-slope transport features and signatures of interaction-driven incompressibility. Because the Hall conductivity is not fully quantized and the longitudinal resistance remains finite, we describe these observations as incipient FCI behavior: transport signatures consistent with a developing fractional Chern insulating state, without claiming a fully developed FCI phase established by complete Hall quantization and vanishing longitudinal resistance. Our results suggest a possible fermiology-driven route toward correlated topological states in dispersive bands in the classical-to-quantum crossover, distinct from the conventional flat-band scenario.

cond-mat.mes-hall↗

arXiv2026

MeSD: Multi-Evidence Self-Distillation for VideoLLM

While reinforcement learning with verifiable rewards provides reliable outcome supervision for VideoLLMs, sequence-level rewards offer limited token-level guidance. On-policy self-distillation addresses this limitation by conditioning a self-teacher on privileged information to provide dense token-level supervision. However, aggregating heterogeneous evidence within a single teacher context obscures cross-evidence agreement and conflict. A further challenge lies in determining whether teacher guidance should refine reward-based updates or provide corrective supervision for failed trajectories. To address these issues, we propose MeSD, a multi-evidence self-distillation framework for VideoLLMs. MeSD constructs three evidence-conditioned teachers with shared parameters, using the ground-truth answer as a common semantic context while separately incorporating temporal and spatial evidence. Given the same student-generated prefixes, MeSD evaluates evidence-specific preferences relative to the Answer Teacher and fuses teacher-common preferences with gated teacher-specific residuals. Furthermore, MeSD introduces Verification-Guided Optimization to classify trajectories as Success, Failure, or Indeterminate. For Success and Indeterminate trajectories, MeSD refines token-level advantage magnitudes while preserving reward-derived signs. For verified failure trajectories that contain the required evidence, MeSD applies failure-conditioned distillation, using reverse-KL correction toward the fused distribution. Experiments on multiple video benchmarks demonstrate consistent gains over reinforcement learning and self-distillation baselines.

cs.CV↗

arXiv2026

An Evaluation of the Semantic Understanding Capabilities of Large Language Models for Web Attack Payloads

Computer vision services delivered through Web interfaces and APIs process textual requests for image-resource acquisition, inference-task configuration, and result management, making Web attack-payload analysis relevant to their deployment security. Large language models (LLMs) can identify payload types and explain attack intent. However, existing studies generally treat payload analysis as a single-layer classification task and lack both a systematic assessment of how deeply LLMs understand payloads and an evaluation benchmark dedicated to the depth of semantic understanding of Web attack payloads. We construct PayloadSemBench, a four-layer semantic evaluation benchmark that operationalizes payload understanding across measurable tasks and comprises 240 payloads. Its ground truth was established through two rounds of anchor calibration and re-verified by a fourth independent expert. Two experiments, a semantic-understanding benchmark and an analysis mapping semantic understanding to detection performance, yielded three main findings: (1) type identification and intent understanding were generally strong, whereas severity assessment was the principal weakness; (2) the effects of obfuscation varied across models and layers, with intent explanation and reconstruction of specific obfuscation techniques more susceptible to degradation, while performance did not degrade synchronously across all layers; and (3) semantic understanding and detection decisions were partially decoupled, with only 12.5% to 50% of missed detections attributable to semantic-understanding failures. External re-evaluation on an independent 180-record dataset comprising production WAF alert streams and real application requests reproduced the non-uniform four-layer capability profile and the layer-specific differences on obfuscated payloads.

cs.CR↗

arXiv2026

Multidimensional Resource Scheduling with Small Demands

We study multidimensional resource scheduling. Each job $i$ has a $d$-dimensional resource-demand vector $v_i$ and a processing time $s_i$. The scheduler assigns a start time to each job, subject to the constraint that, at every time, the total demand of the jobs being processed does not exceed $1$ in any resource dimension. The objective is to minimize the makespan. We focus on the regime in which every individual resource demand is small. We ask whether the favorable \emph{small-vector phenomenon} known for multidimensional vector packing, which corresponds to the special case of unit processing times, extends to jobs with heterogeneous processing times. The key difficulty is that arbitrary processing times create temporal interactions across multiple duration scales: a long job may overlap many shorter jobs, while feasibility must be maintained throughout every job's execution interval. We prove that the small-vector phenomenon persists in this temporal setting. For any $0<ε<1/4$, after normalizing the maximum processing time to $1$, if every coordinate of every demand vector is at most $O(ε^2/\log(d/ε))$, we give a randomized offline algorithm that produces a schedule with expected makespan at most $ (1+6ε)\mathrm{OPT}+3$. Thus, sufficiently small resource demands admit asymptotically near-optimal schedules in arbitrary dimension, despite heterogeneous processing times. We also obtain constant competitive ratios in the online setting. Let $T$ denote the ratio between the maximum and minimum processing times. If every coordinate is at most $O(1/(\log d\log T))$, we give a randomized $O(1)$-competitive algorithm, with a competitive ratio independent of both $d$ and $T$. We further derandomize our approach, obtaining a deterministic $O(1)$-competitive algorithm under a comparable smallness assumption.

cs.DS↗

arXiv2026

Codoku: Renewable Program-Reasoning Challenges for Frontier Coding Agents

Existing program-reasoning benchmarks ask large language models to predict a program's behavior on a given input. Coding agents break two assumptions on which these benchmarks rest: an agent can recover the answer by executing the program instead of reasoning about it, and fixed task sets drawn from existing programs are increasingly exposed to contamination, yet costly to renew. We introduce Codoku (code sudoku), a renewable benchmark in which a solver fills typed cells in a partial program to satisfy global static and dynamic constraints, such as a prescribed control-flow graph and execution path. Because a partial program cannot be executed and valid fillings are sparse in an exponentially large space of interdependent choices, neither tool use nor enumeration can substitute for program reasoning. Puzzles are synthesized from scratch via semantic reification, so fresh puzzles of controllable complexity can be generated on demand, each with a witness that guarantees solvability. We evaluate five frontier models on 300 puzzles through a coding agent free to use any tool within a fixed budget. Small puzzles already challenge open-weight models, whereas even proprietary models solve only about half of the large ones. Codoku thus offers a renewable testbed for program reasoning that can keep pace with rapidly improving coding agents. GitHub: https://github.com/connglli/Codoku.

cs.SE↗