arXiv subjects

Fei Han

Publications and source records attributed to Fei Han.

At least 19 recordsLinked to original sources

arXiv2026

Brane space localization and genus-0 T-duality with $H$-flux

We construct a geometric theory on the brane space $S^2M=C^\infty(S^2,M)$ associated to a gerbe with connection on $M$. Normalized logarithmic holonomy around the latitude circles defines an average holonomy line bundle on the brane space, whose connection and averaged flux give an equivariantly flat superconnection. We prove that the completed periodic exotic twisted cohomology defined by this superconnection localizes to the twisted cohomology of $M$. This {\bf brane space localization} happens in two stages: circle localization lands on the free path space of $M$, and contracting the paths, after a correction of the twisting form, brings us back to $M$. Parallel transport along the meridians lets us average gerbe module connections, after which the circular iterated-integral construction applies. The resulting {\bf twisted brane Bismut-Chern forms} restrict to the twisted Chern forms on $M$; their variation is described by {\bf twisted brane Bismut-Chern-Simons forms.} For a T-dual pair, transporting the classical Hori morphisms through localization yields {\bf graded genus-0 T-duality} on brane cohomology. The composite of each morphism with its dual is minus the Euler operator in the level variable, so these morphisms are isomorphisms at every nonzero level. At level one these morphisms are compatible with the brane character and twisted $K$-theory T-duality.

hep-th↗

arXiv2026

Interior curvature estimates for graphical curvature quotient equations

We prove interior curvature estimates for admissible graphical solutions of curvature quotient equations in the range $3\leq k<n$ on the full Gårding cone $Γ_k$. No convexity, semiconvexity, or $Γ_{k+1}$-admissibility assumption is required. The proof combines a new quantitative compression inequality, a two-surface comparison and doubling argument, and a Pogorelov estimate.

math.AP↗

arXiv2026

Embracing Flow Unsteadiness: A High-Throughput Learning Platform Enables Vortex-Exploiting Bioinspired Propulsion

Biological swimmers and flyers exploit unsteady vortices for propulsion, whereas engineered vehicles usually suppress them as disturbances. Learning such flow exploitation in machines is difficult because real-fluid interaction data are scarce and unstructured exploration is unstable in high-dimensional, history-dependent flows. Here we present REEF, a co-designed physical-learning framework that integrates SHOAL, an eight-channel high-throughput array for real fluid--structure interaction, with V-STAR, a staged algorithm that converts these interactions into policies through imitation, offline internalization, and online adaptation. Across lift-based, drag-based, and momentum-jet propulsors, REEF expands the attainable force envelope to more than twice that of parameterized search. Particle image velocimetry shows that these gains arise from coordinated vortex formation, growth, and force projection, rather than refinement of a fixed motion-to-force mapping. Force-trained policies transfer zero-shot to free-moving robots whose body motion changes the surrounding flow, suggesting that REEF learns transferable wake-coupling principles for embodied propulsion in unsteady fluids.

physics.flu-dyn↗

arXiv2026

An improved bond-associated peridynamic model and its adaptive coupling with CCM for fracture analysis

This paper reformulates the correction factor in the force-state of the bond-associated peridynamic (BAPD) model. The reformulation is established from the strain energy density equivalence between the BAPD model and the classical continuum mechanics (CCM) model at a material point. With the FEM solution taken as the reference, the proposed correction factor improves the accuracy of the BAPD solution in this study. Furthermore, a BAPD-CCM coupled model is developed based on the above energy density equivalence, and a ``Morphing" function is introduced to achieve a smooth transition between the two models. For time integration, explicit schemes are adopted for both quasi-static and dynamic problems. In the spatial discretization, the CCM model is discretized by elements, whereas the BAPD model is discretized by particles. The solution accuracy of the coupled model is validated by comparison with the FEM solution and by evaluating the $L^{2}$ norm, the $H^{1}_{\mathrm{semi}}$, and the energy norm of the displacement error. Two- and three-dimensional numerical examples show that the proposed model has higher computational efficiency. For example, in the Mode I crack propagation problem, its computational cost is reduced by more than 72\% compared with that of the pure BAPD model, and the predicted crack patterns agree with experimental results.

cs.CE↗

arXiv2026

Long SKILL Compliance as Logical Reasoning: Closure-Grounded Detection with Scaling-Guided On-Policy Distillation

The increasing complexity of enterprise business scenarios has promoted the widespread adoption of long SKILL documents in agent systems, posing new challenges for compliance detection: large models incur substantial inference costs, while small models may fail to maintain detection accuracy. To address this gap, we propose SkillCDG, a graph-based framework for long SKILL compliance detection. SkillCDG represents complex business policies as a two-layer constraint dependency graph, where the upper layer indexes SKILL descriptions for scenario routing and the lower layer captures dependencies among atomic constraints within each SKILL. During inference, two-level retrieval followed by dependency closure supports compliance judgment and source traceability. We comprehensively evaluate the framework on three enterprise datasets and two controlled public benchmark variants. Experimental results demonstrate that SkillCDG outperforms baseline methods by up to 12.8 percentage points in detection F1 score, while reducing token consumption by a maximum 64.3\%. Moreover, we further investigate the inherent relationships among policy-graph complexity, model scale, and detection performance. Comparative experiments conducted on four checkpoints from a single model family validate a concise and effective scaling trend: end-to-end detection correctness exhibits a complexity-differentiated scaling pattern, and the complexity metric derived from the constraint dependency graph can effectively quantify instance difficulty and the performance improvement potential of models. Leveraging this insightful scaling trend, we conduct adaptive training sample selection and adopt on-policy distillation to efficiently enhance the compliance detection capability of small-scale models.

cs.AI↗

arXiv2026

Moduli Spaces of Connections and B-fields from T-duality with H-flux

We study the geometry of mixed fields, consisting of a connection and a $B$-field on a principal circle bundle over a Riemann surface, from the perspective of gauge theory and T-duality. Motivated by the foundational work of Atiyah--Bott and Segal, we introduce a twisted Yang--Mills functional whose critical locus, in the flat case, is governed by the simultaneous vanishing of the curvature and the $H$-flux. We show that the gauge group is a semi-direct product of abelian groups parametrised by an integer $λ$. The moduli spaces are constructed by presymplectic reduction and shown to be Heisenberg contact manifolds for $λ\neq 0$, whose topology we characterise completely. We show that T-duality preserves the twisted Yang--Mills functional and acts on the configuration space of flat mixed fields. We identify the subgroups of gauge transformations that are compatible with the T-duality map and describe the induced action on the singular quotient of T-dualizable flat mixed fields. Precisely at $λ=1$ does T-duality descend to an involutive contactomorphism of the moduli space.

hep-th↗

arXiv2026

AutoIQ: An Ensemble Framework for Automatic Assessment of Geometric Distortion in Prostate Diffusion-Weighted Imaging

Geometric distortion in prostate diffusion-weighted imaging (DWI) can impair lesion localization and reduce the reliability of MRI-based clinical assessment. We propose AutoIQ, an ensemble machine learning framework for automatic quantification and classification of DWI geometric distortion severity. A total of 140 retrospective prostate biparametric MRI examinations were analyzed, including 33 scans with severe distortion requiring repeat acquisition and 107 scans with acceptable distortion based on expert radiologist assessment. AutoIQ combines two complementary distortion quantification strategies: a segmentation-based method measuring prostate boundary mismatch between T2-weighted imaging (T2WI) and DWI, and a registration-based method estimating deformation magnitude after DWI-to-T2WI alignment. The resulting distortion scores were used to train individual classifiers and a logistic-regression ensemble model. Both computational methods significantly differentiated severe from acceptable distortion cases (p < 0.001). On an independent test set, the ensemble model achieved an accuracy of 0.95, F1-score of 0.93, and AUC of 0.98, outperforming individual models. These results suggest that AutoIQ can provide automated, quantitative quality assessment for prostate DWI and may help identify scans that require repeat acquisition.

eess.IV↗

arXiv2026

Differential Models for the Anderson Dual to Twisted $\mathrm{Spin}^c$-Bordism and a Twisted Anomaly Map

We construct differential models for degree-3 twisted $\mathrm{Spin}^c$-bordism and for its Anderson dual. The model for the differential Anderson dual is based on the framework of Yamashita--Yonekura. Using these differential models, we define a twisted anomaly map from differential twisted $K$-theory with inverse twist to the differential Anderson dual of twisted $\mathrm{Spin}^c$-bordism. The construction is described geometrically in terms of bundle gerbes, gerbe modules, and reduced eta-invariants of Dirac operators associated to the twisted data. Conceptually, this map is expected to be related to the anomalies of twisted $1|1$-dimensional supersymmetric field theories, in line with the perspectives of Stolz--Teichner and Freed--Hopkins.

math.AT↗

arXiv2026

The Quantization Trap: Breaking Linear Scaling Laws in Multi-Hop Reasoning

Neural scaling laws provide a predictable recipe for AI advancement: reducing numerical precision should linearly improve computational efficiency and energy profile ($E \propto \mathrm{bits}$). In this paper, we demonstrate that this scaling law breaks in the context of multi-hop reasoning. We reveal a 'quantization trap' where reducing precision from 16-bit to 8/4-bit paradoxically increases net energy consumption while degrading reasoning accuracy. We provide a rigorous theoretical decomposition that attributes this failure to hardware casting overhead, the hidden latency cost of dequantization kernels, which becomes a dominant bottleneck in sequential reasoning chains, as well as to a sequential energy amortization failure. As a result, scaling law breaking is unavoidable in practice. We formalize a Critical Model Scale $N^*$ that predicts when the trap dissolves or deepens as a function of model size, batch size, and hardware configuration, validated across a 120$\times$ range (0.6B--72B) on six GPU architectures. Our findings suggest that the industry's "smaller-is-better" heuristic is mathematically counterproductive for complex reasoning tasks.

cs.AI↗

arXiv2026

Omni-LIVO: Robust RGB-Colored Multi-Camera Visual-Inertial-LiDAR Odometry via Photometric Migration and ESIKF Fusion

Wide field-of-view (FoV) LiDAR sensors provide dense geometry across large environments, but existing LiDAR-inertial-visual odometry (LIVO) systems generally rely on a single camera, limiting their ability to fully exploit LiDAR-derived depth for photometric alignment and scene colorization. We present Omni-LIVO, a tightly coupled multi-camera LIVO system that leverages multi-view observations to comprehensively utilize LiDAR geometric information across extended spatial regions. Omni-LIVO introduces a Cross-View direct alignment strategy that maintains photometric consistency across non-overlapping views, and extends the Error-State Iterated Kalman Filter (ESIKF) with multi-view updates and adaptive covariance. The system is evaluated on public benchmarks and our custom dataset, showing improved accuracy and robustness over state-of-the-art LIVO, LIO, and visual-inertial SLAM baselines. Code and dataset will be released upon publication.

cs.RO↗

arXiv2026

Causal World Modeling for Robot Control

This work highlights that video world modeling, alongside vision-language pre-training, establishes a fresh and independent foundation for robot learning. Intuitively, video world models provide the ability to imagine the near future by understanding the causality between actions and visual dynamics. Inspired by this, we introduce LingBot-VA, an autoregressive diffusion framework that learns frame prediction and policy execution simultaneously. Our model features three carefully crafted designs: (1) a shared latent space, integrating vision and action tokens, driven by a Mixture-of-Transformers (MoT) architecture, (2) a closed-loop rollout mechanism, allowing for ongoing acquisition of environmental feedback with ground-truth observations, (3) an asynchronous inference pipeline, parallelizing action prediction and motor execution to support efficient control. We evaluate our model on both simulation benchmarks and real-world scenarios, where it shows significant promise in long-horizon manipulation, data efficiency in post-training, and strong generalizability to novel configurations. The code and model are made publicly available to facilitate the community.

cs.CV↗

arXiv2026

Swimming Under Constraints: A Safe Reinforcement Learning Framework for Quadrupedal Bio-Inspired Propulsion

Bio-inspired aquatic propulsion offers high thrust and maneuverability but is prone to destabilizing forces such as lift fluctuations, which are further amplified by six-degree-of-freedom (6-DoF) fluid coupling. We formulate quadrupedal swimming as a constrained optimization problem that maximizes forward thrust while minimizing destabilizing fluctuations. Our proposed framework, Accelerated Constrained Proximal Policy Optimization with a PID-regulated Lagrange multiplier (ACPPO-PID), enforces constraints with a PID-regulated Lagrange multiplier, accelerates learning via conditional asymmetric clipping, and stabilizes updates through cycle-wise geometric aggregation. Initialized with imitation learning and refined through on-hardware towing-tank experiments, ACPPO-PID produces control policies that transfer effectively to quadrupedal free-swimming trials. Results demonstrate improved thrust efficiency, reduced destabilizing forces, and faster convergence compared with state-of-the-art baselines, underscoring the importance of constraint-aware safe RL for robust and generalizable bio-inspired locomotion in complex fluid environments.

cs.RO↗

arXiv2026

Elliptic Chern Characters and Elliptic Atiyah--Witten Formula

Let $G$ be a compact, connected, and simply connected Lie group. A principal $G$-bundle over a manifold $X$, equipped with a connection, together with a positive-energy representation of the loop group $LG$, gives rise to a circle-equivariant gerbe module on the free loop space $LX$. From this data we construct the elliptic Chern character on $LX$, and a refinement, the elliptic Bismut--Chern character, on the double loop space $L^2X$. Generalizing the classical Atiyah--Witten formula from the free loop space $LX$ to the double loop space $L^2X$, we establish an elliptic Atiyah--Witten formula. The elliptic holonomy on $L^2X$ is defined by $τ$-deformed equivariant twisted parallel transport on $LX$. We show that the four Pfaffian sections, corresponding to the four spin structures on an elliptic curve, are identified with the four elliptic holonomies arising from the four virtual level-one positive-energy representations when $G=\mathrm{Spin}(2n)$. These constructions are intimately connected to the moduli of $G_{\mathbb{C}}$-bundles over elliptic curves and conformal blocks in the context of Chern--Simons gauge theory.

math.DG↗

arXiv2026

Higher Satisfaction, Lower Cost: A Technical Report on How LLMs Revolutionize Meituan's Intelligent Interaction Systems

Enhancing customer experience is essential for business success, particularly as service demands grow in scale and complexity. Generative artificial intelligence and Large Language Models (LLMs) have empowered intelligent interaction systems to deliver efficient, personalized, and 24/7 support. In practice, intelligent interaction systems encounter several challenges: (1) Constructing high-quality data for cold-start training is difficult, hindering self-evolution and raising labor costs. (2) Multi-turn dialogue performance remains suboptimal due to inadequate intent understanding, rule compliance, and solution extraction. (3) Frequent evolution of business rules affects system operability and transferability, constraining low-cost expansion and adaptability. (4) Reliance on a single LLM is insufficient in complex scenarios, where the absence of multi-agent frameworks and effective collaboration undermines process completeness and service quality. (5) The open-domain nature of multi-turn dialogues, lacking unified golden answers, hampers quantitative evaluation and continuous optimization. To address these challenges, we introduce WOWService, an intelligent interaction system tailored for industrial applications. With the integration of LLMs and multi-agent architectures, WOWService enables autonomous task management and collaborative problem-solving. Specifically, WOWService focuses on core modules including data construction, general capability enhancement, business scenario adaptation, multi-agent coordination, and automated evaluation. Currently, WOWService is deployed on the Meituan App, achieving significant gains in key metrics, e.g., User Satisfaction Metric 1 (USM 1) -27.53% and User Satisfaction Metric 2 (USM 2) +25.51%, demonstrating its effectiveness in capturing user needs and advancing personalized service.

cs.CL↗

arXiv2025

Full-bandwidth, continuous, and grayscale 3D nanolithography via line-illumination temporal focusing of ultrafast lasers

Achieving fast and continuous fabrication of large-scale complex 3D structures is key to unlocking industrial-scale adoption of two-photon lithography (TPL). Despite substantial improvement in peak optical patterning rates enabled by recent parallel exposure strategies, the practical fabrication rate of TPL for large structures remains low. This gap is primarily attributed to the mismatched bandwidth among toolpath generation, data transferring, and laser patterning, and the stop-and-go operation for part stitching etc. Here, we present a line-illumination temporal focusing TPL (Line-TF TPL) solution that, for the first time, demonstrates true continuous 3D nanolithography with full-bandwidth data streaming, grayscale voxel tuning, and cost-effective large-scale fabrication capability. To achieve the goal, we use a digital micromirror device (DMD) to temporally focus femtosecond laser pulses into a programmable line with enhanced 3D resolution, pixel-level grayscale control, and a high-refresh rate (>10 kHz), realizing continuous fabrication at a hardware-limited maximum rate. Specifically, we fabricated centimeter-scale 3D structures with sub-diffraction features down to 75 nm laterally and 99 nm axially. Our method eliminates stitching defects by continuous scanning and grayscale stitching; and provides real-time pattern streaming at a bandwidth that is one order of magnitude higher than previous TPL systems. The line-scanning strategy also substantially lowers the pulse-energy requirement, hence the cost for parallel TPL; and maximizes the machine uptime through continuous operation, both of which are critical metrics for industrialization. Finally, we demonstrated centimeter-scale artworks, fine 3D features, and complex miniaturized optics, revealing the Line-TF TPL's large-scale application potential in photonic packaging, metamaterial discovery, and biomedicine.

physics.optics↗

arXiv2025

An Almost Flat Spin$^c$ Manifold Bounds

We prove that every almost flat spin^$c$ manifold bounds a compact orientable manifold, thereby settling, in the spin^$c$ case, a long-standing conjecture of Farrell--Zdravkovska and S. T. Yau.

math.AT↗

arXiv2025

Loop Hori Formulae for T-duality and Twisted Bismut-Chern Character

The main purpose of this paper is to establish the loop space formulation of T-duality in the presence of background flux. In particular, we construct a loop space analogue of the Hori formula, termed \textbf{the loop Hori map}, and demonstrate that it induces a quasi-isomorphism between the exotic twisted equivariant cohomologies on the free loop spaces of the T-dual sides. Spacetime, when viewed as the constant loops, is a submanifold of loop space. The duality that we prove on loop space restricts to the T-duality with $H$-flux on spacetime. This significantly refines the earlier work of the authors in 2015 where T-duality was established after localisation to the base space. The construction of the loop Hori map is an application of our generalization of the Bismut--Chern character in 2015, originally introduced in the loop space interpretation of the Atiyah--Singer index theorem by Atiyah--Witten and Bismut.

hep-th↗

arXiv2025

A new definition of peridynamic damage for thermo-mechanical fracture modeling

A thermo-mechanical fracture modeling is proposed to address thermal failure issues, where the temperature field is calculated by a heat conduction model based on classical continuum mechanics (CCM), while the deformation field with discontinuities is calculated by the peridynamic (PD) model. The model is calculated by a CCM/PD alternating solution based on the finite element discretization, which ensures the calculation accuracy and facilitates engineering applications. The original PD model defines damage solely based on the number of broken bonds in the vicinity of the material point, neglecting the distribution of these bonds. To address this limitation, a new definition of the PD damage accounting for both the number of broken bonds and their specific distribution is proposed. As a result, damage in various directions can be captured, enabling more realistic thermal fracture simulations based on a unified mesh discretization. The effectiveness of the proposed model is validated by comparing numerical examples with analytical solutions. Moreover, simulation results of quasi-static and dynamic crack propagation demonstrate the model's ability to aid in understanding the initiation and propagation mechanisms of complex thermal fractures.

cs.CE↗