arXiv subjects

Miao Wang

Publications and source records attributed to Miao Wang.

At least 19 recordsLinked to original sources

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

Learning Explicit Physical Parameter Control and Benchmarking for Video Generation

Recent advances in image-to-video generation have improved visual realism, making physically grounded and controllable dynamics an important step toward future world simulation. Current models often generate plausible motion, but it is not reliably governed by explicit physical causes, and instance-level constraints can leak or become entangled in multi-object interactions. We attribute this gap to two missing pieces: large-scale, fine-grained physical parameterization, and model designs that correctly bind physical attributes to instances and emphasize dynamics over appearance. To bridge this gap, we introduce PhyParam-Dataset, an interaction-centric collection of 130K physically simulated videos with dense physical parameterization, including force vectors, object material properties, and environmental constants across five representative rigid-body motion types. Built on this data, we present PhyParam, a physics-guided image-to-video diffusion model that conditions on object-level forces, masses, friction, restitution, and scene-level gravity via a lightweight physical-attention routing mechanism, and further improves motion learning with semantic-structural feature-space supervision. We also establish PhyParam-Bench, a benchmark for physical-law consistency in image-to-video generation, with a multi-level protocol evaluating temporal dynamics, spatial stability, and semantic--physical alignment. Experiments show that PhyParam improves physical consistency while maintaining high visual fidelity, advancing explicit rigid-body physical-parameter control for image-to-video generation. We will publicly release the dataset, benchmark, and code to support future research.

cs.CV↗

arXiv2026

Disturbance-Aware Flight for Aerial Robots in Narrow Space

Autonomous flight of aerial robots in narrow space remains challenging due to strong aerodynamic disturbances and limited flying space. Existing approaches mainly address aerodynamic disturbances at the control level, while motion planning typically relies on geometric constraints and fixed speed limits, leading to conservative or unsafe behaviors in confined environments. This paper presents a disturbance-aware planning and control framework (DAPCF) that integrates online disturbance estimation into the planning-control loop for quadrotor flight in narrow space. First, the dual-loop observers estimate 6-degree-of-freedom disturbance forces and torques in real time based on odometry and motor speed measurements. Then, a disturbance risk function is introduced that adaptively modulates the reference speed of the planner based on disturbance estimation, reducing velocity when disturbances exceed a threshold and restoring it under low-disturbance conditions. Finally, a motor-dynamics-based nonlinear model predictive controller (MDNMPC) with disturbance compensation is designed to ensure robust trajectory tracking under perturbed conditions. Experiments demonstrate that a quadrotor with a diagonal length of 0.39~m can traverse straight, sloped, and curved tunnels as narrow as 0.6~m, outperforming human pilots in both success rate and flight efficiency.

cs.RO↗

arXiv2026

Configuration-Induced Passive Self-Rotation for Perception-Enhanced Autonomous Flight

Autonomous flight in confined and cluttered environments is fundamentally limited by the restricted field of view (FoV) of onboard sensors. Passive self-rotation expands sensing coverage without additional sensors but introduces a tradeoff between swept-FoV refresh rate and flight performance. This letter presents a configuration-induced passively self-rotating tricopter for perception-enhanced autonomous flight. Firstly, the rear-arm configuration parameter is exploited to regulate the passive self-rotation operating point, providing an airframe-level mechanism for balancing swept-FoV refresh rate and flight performance. Secondly, a hierarchical autonomy framework integrating planning and control is developed to enable agile and robust autonomous flight under continuous passive self-rotation. For waypoint-based inspection, guide-point replanning is further used to improve task-level coverage. Extensive real-world experiments, including high-speed trajectory tracking, disturbance-rejection tests, and autonomous navigation in representative cluttered environments, demonstrate the effectiveness of the proposed approach for perception-enhanced autonomous flight.

cs.RO↗

arXiv2026

IMWM: Intuition Models Complement World Models for Latent Planning

Planning with a learned latent world model is a promising route to control from raw pixels, but a strong world model alone is not enough. We show this experimentally: even with a perfect world model (operationalized by replacing the learned forward predictor with an idealized rollout of the true environment dynamics), a finite-budget sample-based planner still fails on some tasks, indicating that the bottleneck can lie in search rather than in world-model accuracy. Motivated by this gap, we propose IMWM (Intuition Model + World Model), which pairs the world model with an intuition model trained from demonstrations to recognize promising actions. The two models collaborate through three lightweight components: (i) Retrieval Initialization, which initializes the planner's action proposal from a retrieved demonstration; (ii) Hybrid Cost, which combines the intuition score with the world-model rollout cost; and (iii) a Reliability Gate, which adjusts how much the planner trusts intuition in each setting. Across four pixel-based goal-reaching tasks (Two-Room, Reacher, Push-T, and OGBench-Cube), IMWM has higher mean success than the world-model-only planner on all four, with the largest gains on Two-Room (99.2%, +11.5 percentage points) and OGBench-Cube (94.7%, +28.5 percentage points).

cs.LG↗

arXiv2026

Reward-Decomposed Reinforcement Learning for Immersive Video Role-Playing

Text-based role-playing models can imitate character styles, but often fail to capture scene atmosphere and evolving tension, which are crucial for immersive applications such as VR games and interactive narratives. We study video-grounded role-playing dialogue and introduce EBM-RL (Eye--Brain--Mouth Reinforcement Learning), a decoupled GRPO-based framework that separates observation ( ), reasoning ( ), and utterance generation ( ). This design mimics the human See-Think-Speak process, enabling the model to ground dialogue in visual perception before reasoning and response generation. To optimize this See-Think-Speak process, EBM-RL integrates complementary rewards for scene--text alignment, perceptual--cognitive utility, answer faithfulness, and format consistency. Extensive experiments show that EBM-RL substantially outperforms text-only role-playing baselines and larger-scale vision-language models on our immersive role-playing benchmark, improving both visual-atmosphere consistency and character authenticity. Moreover, EBM-RL demonstrates strong zero-shot transfer to out-of-domain VideoQA benchmarks without additional fine-tuning. We also release an open-source dataset for video-grounded role-playing dialogue.

cs.AI↗

arXiv2026

Rotor-Failure-Aware Quadrotors Flight in Unknown Environments

Rotor failures in quadrotors may result in high-speed rotation and vibration due to rotor imbalance, which introduces significant challenges for autonomous flight in unknown environments. The mainstream approaches against rotor failures rely on fault-tolerant control (FTC) and predefined trajectory tracking. To the best of our knowledge, online failure detection and diagnosis (FDD), trajectory planning, and FTC of the post-failure quadrotors in unknown and complex environments have not yet been achieved. This paper presents a rotor-failure-aware quadrotor navigation system designed to mitigate the impacts of rotor imbalance. First, a composite FDD-based nonlinear model predictive controller (NMPC), incorporating motor dynamics, is designed to ensure fast failure detection and flight stability. Second, a rotor-failure-aware planner is designed to leverage FDD results and spatial-temporal joint optimization, while a LiDAR-based quadrotor platform with four anti-torque plates is designed to enable reliable perception under high-speed rotation. Lastly, extensive benchmarks against state-of-the-art methods highlight the superior performance of the proposed approach in addressing rotor failures, including propeller unloading and motor stoppage. The experimental results demonstrate, for the first time, that our approach enables autonomous quadrotor flight with rotor failures in challenging environments, including cluttered rooms and unknown forests.

cs.RO↗

arXiv2026

Covering Hypercube $mB^n$

The Alon--F\"uredi theorem determines the minimum number of hyperplanes needed to cover every nonzero vertex of the Boolean cube while avoiding the origin. We study hyperplane coverings with multiplicity for the generalized hypercube $mB^n=\{0,1,\ldots,m\}^n$. Our main result is a polynomial multiplicity theorem extending the theorem of Sauermann and Wigderson to $mB^n$, with sharp degree bounds for polynomials having prescribed high-order zeros. We apply this result to obtain lower and upper bounds for hyperplane coverings with multiplicity and determine the exact minimum number of hyperplanes for coverings with multiplicities $k=1,2$. In the special case of the Boolean cube, our constructions improve the upper bound of Clifton and Huang for certain ranges of the parameters.

math.CO↗

arXiv2026

Higher-Order Corrections to Scrambling Dynamics in Brownian Spin SYK Models

We investigate operator growth in a Brownian spin Sachdev--Ye--Kitaev (SYK) model with random all-to-all interactions, focusing on the full operator-size distribution. For Hamiltonians containing interactions of order two up to $L$, we derive a closed master equation for the Pauli-string expansion coefficients and recast their dynamics into a generating-function formulation suitable for the large-$N$ limit. This approach allows us to diagonalize the leading-order evolution operator explicitly and obtain exact solutions for arbitrary initial operator distributions, including the effects of decoherence. Going beyond leading order, we develop a systematic $1/N$ expansion that captures higher-order corrections to the operator-size dynamics and the late-time behavior. Our results demonstrate that higher-order effects play a crucial role in operator scrambling and that the full operator-size distribution provides a more refined probe of quantum chaos in Brownian and open quantum systems.

quant-ph↗

arXiv2026

Trifuse: Enhancing Attention-Based GUI Grounding via Multimodal Fusion

GUI grounding maps natural language instructions to the correct interface elements, serving as the perception foundation for GUI agents. Existing approaches predominantly rely on fine-tuning multimodal large language models (MLLMs) using large-scale GUI datasets to predict target element coordinates, which is data-intensive and generalizes poorly to unseen interfaces. Recent attention-based alternatives exploit localization signals in MLLMs attention mechanisms without task-specific fine-tuning, but suffer from low reliability due to the lack of explicit and complementary spatial anchors in GUI images. To address this limitation, we propose Trifuse, an attention-based grounding framework that explicitly integrates complementary spatial anchors. Trifuse integrates attention, OCR-derived textual cues, and icon-level caption semantics via a Consensus-SinglePeak (CS) fusion strategy that enforces cross-modal agreement while retaining sharp localization peaks. Extensive evaluations on four grounding benchmarks demonstrate that Trifuse achieves strong performance without task-specific fine-tuning, substantially reducing the reliance on expensive annotated data. Moreover, ablation studies reveal that incorporating OCR and caption cues consistently improves attention-based grounding performance across different backbones, highlighting its effectiveness as a general framework for GUI grounding.

cs.AI↗

arXiv2026

SC-MAS: Constructing Cost-Efficient Multi-Agent Systems with Edge-Level Heterogeneous Collaboration

Large Language Model (LLM)-based Multi-Agent Systems (MAS) enhance complex problem solving through multi-agent collaboration, but often incur substantially higher costs than single-agent systems. Recent MAS routing methods aim to balance performance and overhead by dynamically selecting agent roles and language models. However, these approaches typically rely on a homogeneous collaboration mode, where all agents follow the same interaction pattern, limiting collaboration flexibility across different roles. Motivated by Social Capital Theory, which emphasizes that different roles benefit from distinct forms of collaboration, we propose SC-MAS, a framework for constructing heterogeneous and cost-efficient multi-agent systems. SC-MAS models MAS as directed graphs, where edges explicitly represent pairwise collaboration strategies, allowing different agent pairs to interact through tailored communication patterns. Given an input query, a unified controller progressively constructs an executable MAS by selecting task-relevant agent roles, assigning edge-level collaboration strategies, and allocating appropriate LLM backbones to individual agents. Experiments on multiple benchmarks demonstrate the effectiveness of SC-MAS. In particular, SC-MAS improves accuracy by 3.35% on MMLU while reducing inference cost by 15.38%, and achieves a 3.53% accuracy gain with a 12.13% cost reduction on MBPP. These results validate the feasibility of SC-MAS and highlight the effectiveness of heterogeneous collaboration in multi-agent systems.

cs.MA↗

arXiv2026

$E$ and $J$ type $\mathcal{N}=(0,2)$ disordered models and higher-spin symmetry

In this work, we investigate the emergence of higher-spin structure in 2d $\mathcal{N}=(0,2)$ disordered models. While previous studies focused on the $J$-type model where the $E$-term in the Fermi multiplet was discarded. We extend the discussion to $\mathcal{N}=(0,2)$ disordered models with $E$-type potential. In terms of (disordered) $\mathcal{N}=(0,2)$ Landau-Ginzburg theory, we establish a duality between two models. By solving the Schwinger-Dyson equations and the ladder kernel matrix for 4-point functions, we verify that the $E$-type model is dynamically equivalent to the $J$-type model in the IR regime. Furthermore, we demonstrate that the $E$-type model also exhibits emergent higher-spin symmetry in certain limits. Our results reveal a larger region of the moduli space of 2D $\mathcal{N}=(0,2)$ disordered theories and provides insights into the holographic transition from finite to tensionless strings that can be diagnosed by the emergence of higher-spin symmetries.

hep-th↗

arXiv2025

Multi-Agent Intelligence for Multidisciplinary Decision-Making in Gastrointestinal Oncology

Multimodal clinical reasoning in the field of gastrointestinal (GI) oncology necessitates the integrated interpretation of endoscopic imagery, radiological data, and biochemical markers. Despite the evident potential exhibited by Multimodal Large Language Models (MLLMs), they frequently encounter challenges such as context dilution and hallucination when confronted with intricate, heterogeneous medical histories. In order to address these limitations, a hierarchical Multi-Agent Framework is proposed, which emulates the collaborative workflow of a human Multidisciplinary Team (MDT). The system attained a composite expert evaluation score of 4.60/5.00, thereby demonstrating a substantial improvement over the monolithic baseline. It is noteworthy that the agent-based architecture yielded the most substantial enhancements in reasoning logic and medical accuracy. The findings indicate that mimetic, agent-based collaboration provides a scalable, interpretable, and clinically robust paradigm for automated decision support in oncology.

cs.AI↗

arXiv2025

Prefrontal scaling of reward prediction error readout gates reinforcement-derived adaptive behavior in primates

Reinforcement learning (RL) enables adaptive behavior across species via reward prediction errors (RPEs), but the neural origins of species-specific adaptability remain unknown. Integrating RL modeling, transcriptomics, and neuroimaging during reversal learning, we discovered convergent RPE signatures - shared monoaminergic/synaptic gene upregulation and neuroanatomical representations, yet humans outperformed macaques behaviorally. Single-trial decoding showed RPEs guided choices similarly in both species, but humans disproportionately recruited dorsal anterior cingulate (dACC) and dorsolateral prefrontal cortex (dlPFC). Cross-species alignment uncovered that macaque prefrontal circuits encode human-like optimal RPEs yet fail to translate them into action. Adaptability scaled not with RPE encoding fidelity, but with the areal extent of dACC/dlPFC recruitment governing RPE-to-action transformation. These findings resolve an evolutionary puzzle: behavioral performance gaps arise from executive cortical readout efficiency, not encoding capacity.

q-bio.NC↗

arXiv2025

Co-TAP: Three-Layer Agent Interaction Protocol Technical Report

This paper proposes Co-TAP (T: Triple, A: Agent, P: Protocol), a three-layer agent interaction protocol designed to address the challenges faced by multi-agent systems across the three core dimensions of Interoperability, Interaction and Collaboration, and Knowledge Sharing. We have designed and proposed a layered solution composed of three core protocols: the Human-Agent Interaction Protocol (HAI), the Unified Agent Protocol (UAP), and the Memory-Extraction-Knowledge Protocol (MEK). HAI focuses on the interaction layer, standardizing the flow of information between users, interfaces, and agents by defining a standardized, event-driven communication paradigm. This ensures the real-time performance, reliability, and synergy of interactions. As the core of the infrastructure layer, UAP is designed to break down communication barriers among heterogeneous agents through unified service discovery and protocol conversion mechanisms, thereby enabling seamless interconnection and interoperability of the underlying network. MEK, in turn, operates at the cognitive layer. By establishing a standardized ''Memory (M) - Extraction (E) - Knowledge (K)'' cognitive chain, it empowers agents with the ability to learn from individual experiences and form shareable knowledge, thereby laying the foundation for the realization of true collective intelligence. We believe this protocol framework will provide a solid engineering foundation and theoretical guidance for building the next generation of efficient, scalable, and intelligent multi-agent applications.

cs.AI↗

arXiv2025

Mono4DEditor: Text-Driven 4D Scene Editing from Monocular Video via Point-Level Localization of Language-Embedded Gaussians

Editing 4D scenes reconstructed from monocular videos based on text prompts is a valuable yet challenging task with broad applications in content creation and virtual environments. The key difficulty lies in achieving semantically precise edits in localized regions of complex, dynamic scenes, while preserving the integrity of unedited content. To address this, we introduce Mono4DEditor, a novel framework for flexible and accurate text-driven 4D scene editing. Our method augments 3D Gaussians with quantized CLIP features to form a language-embedded dynamic representation, enabling efficient semantic querying of arbitrary spatial regions. We further propose a two-stage point-level localization strategy that first selects candidate Gaussians via CLIP similarity and then refines their spatial extent to improve accuracy. Finally, targeted edits are performed on localized regions using a diffusion-based video editing model, with flow and scribble guidance ensuring spatial fidelity and temporal coherence. Extensive experiments demonstrate that Mono4DEditor enables high-quality, text-driven edits across diverse scenes and object types, while preserving the appearance and geometry of unedited areas and surpassing prior approaches in both flexibility and visual fidelity.

cs.CV↗

arXiv2025

Heterogeneity-Oblivious Robust Federated Learning

Federated Learning (FL) remains highly vulnerable to poisoning attacks, especially under real-world hyper-heterogeneity, where clients differ significantly in data distributions, communication capabilities, and model architectures. Such heterogeneity not only undermines the effectiveness of aggregation strategies but also makes attacks more difficult to detect. Furthermore, high-dimensional models expand the attack surface. To address these challenges, we propose Horus, a heterogeneity-oblivious robust FL framework centered on low-rank adaptations (LoRAs). Rather than aggregating full model parameters, Horus inserts LoRAs into empirically stable layers and aggregates only LoRAs to reduce the attack uncover a key empirical observation that the input projection (LoRA-A) is markedly more stable than the output projection (LoRA-B) under heterogeneity and poisoning. Leveraging this, we design a Heterogeneity-Oblivious Poisoning Score using the features from LoRA-A to filter poisoned clients. For the remaining benign clients, we propose projection-aware aggregation mechanism to preserve collaborative signals while suppressing drifts, which reweights client updates by consistency with the global directions. Extensive experiments across diverse datasets, model architectures, and attacks demonstrate that Horus consistently outperforms state-of-the-art baselines in both robustness and accuracy.

cs.LG↗

arXiv2025

Convergence and consensus analysis of a class of best-response opinion dynamics

Opinion dynamics aims to understand how individuals' opinions evolve through local interactions. Recently, opinion dynamics have been modeled as network games, where individuals update their opinions in order to minimize the social pressure caused by disagreeing with others. In this paper, we study a class of best response opinion dynamics introduced by Mei et al., where a parameter $α> 0$ controls the marginal cost of opinion differences, bridging well-known mechanisms such as the DeGroot model ($α= 2$) and the weighted-median model ($α= 1$). We conduct theoretical analysis on how different values of $α$ affect the system's convergence and consensus behavior. For the case when $α> 1$, corresponding to increasing marginal costs, we establish the convergence of the dynamics and derive graph-theoretic conditions for consensus formation, which is proved to be similar to those in the DeGroot model. When $α< 1$, we show via a counterexample that convergence is not always guaranteed, and we provide sufficient conditions for convergence and consensus. Additionally, numerical simulations on small-world networks reveal how network structure and $α$ together affect opinion diversity.

math.DS↗