arXiv subjects

Xiao Li

Publications and source records attributed to Xiao Li.

At least 109 records · Page 6Linked to original sources

arXiv2026

Impurity-induced thermal crossover in fractional Chern insulators

The recent experimental observation of fractional quantum anomalous Hall (FQAH) states in rhombohedral multilayer graphene has attracted significant attention. One of the most intriguing observations is that the FQAH states at various fractional fillings give way to IQAH states as the temperature is lowered. In this work, we propose a mechanism for the appearance of FQAH states within a finite temperature range in a toy model. The model consists of a flat Chern band and impurities, and we analyze the effects of impurities on the system's behavior at finite temperatures. We believe that the crossover may arise from the competition between the energy penalty for thermal excitations and the increase in entropy. We support our theoretical argument with numerical calculations using exact diagonalization. Our results suggest that impurities may play a crucial role in the crossover from the FQAH to IQAH states in rhombohedral pentalayer graphene.

cond-mat.str-el↗

arXiv2026

Bound preserving and mass conservative methods for the nonlocal Cahn-Hilliard equation with the logarithmic Flory-Huggins potential

It is well known that the exponential time differencing (ETD) method has been successfully applied to the classic Cahn-Hilliard equation with double well potential. However, this numerical method can not be extended to the Cahn-Hilliard equation with Flory-Huggins potential directly due to the fact that the the numerical solution may go beyond the physical interval which leads the non-physical solution. In this paper, we develop and analyze first- and second-order numerical schemes for the nonlocal Cahn-Hilliard equation with the classic Flory-Huggins energy potential. In more detail, the ETD method is firstly used to obtain the prediction solution, and then this prediction solution is corrected by the projection method to avoid non-physical solution. The proposed method is shown to preserve bound and mass conservation in discrete settings. In addition, error estimates for the numerical solution are rigorously obtained for both schemes. Extensive numerical tests and comparisons are conducted to demonstrate the performance of the proposed schemes.

math.NA↗

arXiv2026

Bringing Your Portrait to 3D Presence

We present a unified framework for reconstructing animatable 3D human avatars from a single portrait across head, half-body, and full-body inputs. Our method tackles three bottlenecks: pose- and framing-sensitive feature representations, limited scalable data, and unreliable proxy-mesh estimation. We introduce a Dual-UV representation that maps image features to a canonical UV space via Core-UV and Shell-UV branches, eliminating pose- and framing-induced token shifts. We also build a factorized synthetic data manifold combining 2D generative diversity with geometry-consistent 3D renderings, supported by a training scheme that improves realism and identity consistency. A robust proxy-mesh tracker maintains stability under partial visibility. Together, these components enable strong in-the-wild generalization. Trained only on half-body synthetic data, our model achieves state-of-the-art head and upper-body reconstruction and competitive full-body results. Extensive experiments and analyses further validate the effectiveness of our approach.

cs.CV↗

arXiv2026

Hierarchical Long Video Understanding with Audiovisual Entity Cohesion and Agentic Search

Long video understanding presents significant challenges for vision-language models due to extremely long context windows. Existing solutions relying on naive chunking strategies with retrieval-augmented generation, typically suffer from information fragmentation and a loss of global coherence. We present HAVEN, a unified framework for long-video understanding that enables coherent and comprehensive reasoning by integrating audiovisual entity cohesion and hierarchical video indexing with agentic search. First, we preserve semantic consistency by integrating entity-level representations across visual and auditory streams, while organizing content into a structured hierarchy spanning global summary, scene, segment, and entity levels. Then we employ an agentic search mechanism to enable dynamic retrieval and reasoning across these layers, facilitating coherent narrative reconstruction and fine-grained entity tracking. Extensive experiments demonstrate that our method achieves good temporal coherence, entity consistency, and retrieval efficiency, establishing a new state-of-the-art with an overall accuracy of 84.1% on LVBench. Notably, it achieves outstanding performance in the challenging reasoning category, reaching 80.1%. These results highlight the effectiveness of structured, multimodal reasoning for comprehensive and context-consistent understanding of long-form videos.

cs.CV↗

arXiv2026

In-orbit Test of the Weak Equivalence Principle with Atom Interferometry

The Weak Equivalence Principle (WEP) is a central pillar of general relativity. Its precise test with quantum systems in space offers a unique window onto new physics. Here we report the first in-orbit quantum test of the WEP. A dual-species (85Rb/87Rb) atom interferometer is realized aboard the China Space Station. Methods of platform motion suppression, fluorescence detection switching, and two-photon detuning switching are developed to eliminate phase noise and improve measurement accuracy. A test uncertainty of 2.8*10-8 is obtained from 280 days of WEP test data, and a test result of (-3.1+/-4.6)*10-7 is achieved after error estimation. This improves prior atom-interferometric WEP tests in microgravity by three orders of magnitude. This work paves the way for space-borne quantum inertial sensors and their application to future fundamental physics in space.

physics.atom-ph↗

arXiv2026

Beam Prediction Based on Multimodal Large Language Models

Accurate beam prediction is a key enabler for next-generation wireless communication systems. In this paper, we propose a multimodal large language model (LLM)-based beam prediction framework that effectively utilizes contextual information, provided by sensory data including RGB camera images and LiDAR point clouds. To effectively fuse heterogeneous modalities, we design specialized modality encoders together with a beam-guided attention masking mechanism and a high-frequency temporal alignment strategy, enabling robust cross-modal feature integration under dynamic environments. Furthermore, we construct a large-scale multimodal dataset for communication, named Multimodal-Wireless, which covers diverse weather and traffic conditions with high-fidelity ray-tracing labels. Extensive simulation results demonstrate that the proposed approach significantly reduces the reliance on oracle angle-of-departure knowledge and consistently outperforms state-of-the-art multimodal LLM-based beam prediction methods in terms of beam accuracy and communication performance, improving the average Top-1 accuracy to 80.8% and the average normalized gain to 89.1%.

eess.SP↗

arXiv2026

Design and implementation of a high-density sub-nanosecond timing system for a C-band photocathode electron gun test platform

This paper presents the design and implementation of a high-density, deterministic trigger distribution system tailored for the C-band photocathode electron gun test platform at the Southern Advanced Photon Source (SAPS). Implemented within a scalable 6U VME modular architecture, the system achieves high-density integration by consolidating a master controller, clock distribution network, and 80 heterogeneous output channels into a single chassis. This design leverages a high-performance FPGA core combined with custom backplane interconnections to establish a master-slave topology, significantly reducing the system footprint compared to stacked standalone generators. To guarantee timing determinism in high-noise environments, precise placement and timing constraints are applied to the FPGA logic, while optical isolation is employed to mitigate electromagnetic interference. Furthermore, a dual-channel SFP optical signaling architecture enables seamless expansion to 160 synchronized channels. A remote control framework based on a serial server and a virtual machine Input/Output Controller (IOC) facilitates flexible configuration. Performance tests demonstrate adjustable trigger frequencies from 1 Hz to 100 Hz, with delays and pulse widths tunable from 0 to 10 ms at a resolution of 10 ns (or the RF period). The local electrical output exhibits an ultra-low RMS jitter of 6.55 ps (60 ps peak-to-peak). For remote optical distribution, the system maintains a sub-nanosecond RMS jitter of 119.5 ps, with peak-to-peak variation confined to 1 ns due to the combined effects of transceiver optoelectronic conversion (utilizing HFBR-1414T/2412T modules) and fiber transmission. The system has been successfully commissioned and is currently in reliable routine operation, verifying the architecture as a robust, highly integrated, and cost-effective solution for compact accelerator facilities.

physics.acc-ph↗

arXiv2026

Predicting power grid frequency dynamics with invertible Koopman-based architectures

The system frequency is a critical measure of power system stability and understanding, and modeling it are key to ensure reliable power system operations. Koopman-based autoencoders are effective at approximating complex nonlinear data patterns, with potential applications in the frequency dynamics of power systems. However, their non-invertibility can result in a distorted latent representation, leading to significant prediction errors. Invertible neural networks (INNs) in combination with the Koopman operator framework provide a promising approach to address these limitations. In this study, we analyze different INN architectures and train them on simulation datasets. We further apply extensions to the networks to address inherent limitations of INNs and evaluate their impact. We find that coupling-layer INNs achieve the best performance when used in isolation. In addition, we demonstrate that hybrid approaches can improve the performance when combined with suitable INNs, while reducing the generalization capabilities in combination with disadvantageous architectures. Overall, our results provide a clearer overview of how architectural choices influence INN performance, offering guidance for selecting and designing INNs for modeling power system frequency dynamics.

eess.SY↗

arXiv2026

Structure-Aware Multimodal LLM Framework for Trustworthy Near-Field Beam Prediction

In near-field extremely large-scale multiple-input multiple-output (XL-MIMO) systems, spherical wavefront propagation expands the traditional beam codebook into the joint angular-distance domain, rendering conventional beam training prohibitively inefficient, especially in complex 3-dimensional (3D) low-altitude environments. Furthermore, since near-field beam variations are deeply coupled not only with user positions but also with the physical surroundings, precise beam alignment demands profound environmental understanding capabilities. To address this, we propose a large language model (LLM)-driven multimodal framework that fuses historical GPS data, RGB image, LiDAR data, and strategically designed task-specific textual prompts. By utilizing the powerful emergent reasoning and generalization capabilities of the LLM, our approach learns complex spatial dynamics to achieve superior environmental comprehension...

eess.SP↗

arXiv2026

Load-Aware Locomotion Control for Humanoid Robots in Industrial Transportation Tasks

Humanoid robots deployed in industrial environments are required to perform load-carrying transportation tasks that tightly couple locomotion and manipulation. However, achieving stable and robust locomotion under varying payloads and upper-body motions is challenging due to dynamic coupling and partial observability. This paper presents a load-aware locomotion framework for industrial humanoids based on a decoupled yet coordinated loco-manipulation architecture. Lower-body locomotion is controlled via a reinforcement learning policy producing residual joint actions on kinematically derived nominal configurations. A kinematics-based locomotion reference with a height-conditioned joint-space offset guides learning, while a history-based state estimator infers base linear velocity and height and encodes residual load- and manipulation-induced disturbances in a compact latent representation. The framework is trained entirely in simulation and deployed on a full-size humanoid robot without fine-tuning. Simulation and real-world experiments demonstrate faster training, accurate height tracking, and stable loco-manipulation. Project page: https://lequn-f.github.io/LALO/

cs.RO↗

arXiv2026

Improving Channel Estimation via Multimodal Diffusion Models with Flow Matching

Deep generative models offer a powerful alternative to conventional channel estimation by learning complex channel distributions. By integrating the rich environmental information available in modern sensing-aided networks, this paper proposes MultiCE-Flow, a multimodal channel estimation framework based on flow matching and diffusion transformer (DiT). We design a specialized multimodal perception module that fuses LiDAR, camera, and location data into a semantic condition, while treating sparse pilots as a structural condition. These conditions guide a DiT backbone to reconstruct high-fidelity channels. Unlike standard diffusion models, we employ flow matching to learn a linear trajectory from noise to data, enabling efficient one-step sampling. By leveraging environmental semantics, our method mitigates the ill-posed nature of estimation with sparse pilots. Extensive experiments demonstrate that MultiCE-Flow consistently outperforms traditional baselines and existing generative models. Notably, it exhibits superior robustness to out-of-distribution scenarios and varying pilot densities, making it suitable for environment-aware communication systems.

cs.LG↗

arXiv2026

Compose Your Policies! Improving Diffusion-based or Flow-based Robot Policies via Test-time Distribution-level Composition

Diffusion-based models for robotic control, including vision-language-action (VLA) and vision-action (VA) policies, have demonstrated significant capabilities. Yet their advancement is constrained by the high cost of acquiring large-scale interaction datasets. This work introduces an alternative paradigm for enhancing policy performance without additional model training. Perhaps surprisingly, we demonstrate that the composed policies can exceed the performance of either parent policy. Our contribution is threefold. First, we establish a theoretical foundation showing that the convex composition of distributional scores from multiple diffusion models can yield a superior one-step functional objective compared to any individual score. A Grönwall-type bound is then used to show that this single-step improvement propagates through entire generation trajectories, leading to systemic performance gains. Second, motivated by these results, we propose General Policy Composition (GPC), a training-free method that enhances performance by combining the distributional scores of multiple pre-trained policies via a convex combination and test-time search. GPC is versatile, allowing for the plug-and-play composition of heterogeneous policies, including VA and VLA models, as well as those based on diffusion or flow-matching, irrespective of their input visual modalities. Third, we provide extensive empirical validation. Experiments on Robomimic, PushT, and RoboTwin benchmarks, alongside real-world robotic evaluations, confirm that GPC consistently improves performance and adaptability across a diverse set of tasks. Further analysis of alternative composition operators and weighting strategies offers insights into the mechanisms underlying the success of GPC. These results establish GPC as a simple yet effective method for improving control performance by leveraging existing policies.

cs.RO↗

arXiv2026

Meta-cavity Quantum Electrodynamics

Cavity quantum electrodynamics (cQED) harnesses light-matter interactions to produce nonclassical light states. However, a fundamental challenge lies in simultaneously achieving Purcell enhancement and tailored wavefront control within a single cavity, due to conflicting resonator requirements. Here, we overcome this limitation by demonstrating triggered single-photon emission with customizable wavefronts from semiconductor quantum dots embedded in geometric-phase metacavities. These monolithic devices - only 200 nm thick - deliver Purcell-enhanced emission alongside spin-momentum-locked radiation, vortex beams, and holographic patterns. The meta-atom lattice provides high-Q optical confinement, while spatially modulated orientations enable efficient outcoupling of photons with designed states. This work establishes a new paradigm for intrinsically multiplexing metasurface-based wavefront shaping with cQED, enabling high-performance quantum light sources from subwavelength-scale monolithic platforms.

physics.optics↗

arXiv2026

Unveiling Downstream Performance Scaling of LLMs: A Clustering-Based Perspective

The escalating scale and cost of Large Language Models (LLMs) training necessitate accurate pre-training prediction of downstream task performance for comprehensive understanding of scaling properties. This is challenged by: 1) the emergence phenomenon, where unpredictable capabilities appearing suddenly at critical model scales; and 2) uneven task difficulty and inconsistent performance scaling patterns, leading to high metric variability. Current prediction methods lack accuracy and reliability. We propose a Clustering-On-Difficulty (COD) framework for downstream performance prediction. The COD framework clusters tasks by their difficulty scaling features, thereby constructing a more stable and predictable task subset that exhibits well-behaved scaling characteristics with the increase of compute budget. We adopt a performance scaling law to predict cluster-wise performance with theoretical support. Predictable subset performance acts as an intermediate predictor for the full evaluation set. We further derive a mapping function to accurately extrapolate the performance of the subset to the full set. Applied to an LLM with 70B parameters, COD achieved a 1.55\% average prediction error across eight key LLM benchmarks, thus providing actionable insights for scaling properties and training monitoring during LLM pre-training.

cs.CL↗

arXiv2026

Valleytronics in 2D Materials Roadmap

Valleytronics exploits non-equivalent energy extrema in the electronic band structure of crystalline solids -- the valley degree of freedom -- to encode, manipulate, and read out information. The advent of 2D materials, first graphene and then transition-metal dichalcogenides, made valley control practical through optical, electrical, and magnetic routes. This foundation has enabled remarkable progress in recent years spanning established frontiers, such as valley exciton physics and valley Hall effects, as well as emerging directions including lightwave valleytronics, nanophotonic integration, flat-band valleytronics, and spin-valley qubits. In parallel, there are sustained efforts to scale up valleytronic materials and to predict new valleytronic platforms. This Roadmap brings together perspectives from leading experts to chart the key opportunities and challenges at the forefront of 2D material valleytronics. Each section captures a snapshot of progress in a key research area, identifies critical open challenges, and outlines pathways toward future valleytronics breakthroughs.

cond-mat.mes-hall↗

arXiv2026

CharacterFlywheel: Scaling Iterative Improvement of Engaging and Steerable LLMs in Production

This report presents CharacterFlywheel, an iterative flywheel process for improving large language models (LLMs) in production social chat applications across Instagram, WhatsApp, and Messenger. Starting from LLaMA 3.1, we refined models across 15 generations using data from both internal and external real-user traffic. Through continuous deployments from July 2024 to April 2025, we conducted controlled 7-day A/B tests showing consistent engagement improvements: 7 of 8 newly deployed models demonstrated positive lift over the baseline, with the strongest performers achieving up to 8.8% improvement in engagement breadth and 19.4% in engagement depth. We also observed substantial gains in steerability, with instruction following increasing from 59.2% to 84.8% and instruction violations decreasing from 26.6% to 5.8%. We detail the CharacterFlywheel process which integrates data curation, reward modeling to estimate and interpolate the landscape of engagement metrics, supervised fine-tuning (SFT), reinforcement learning (RL), and both offline and online evaluation to ensure reliable progress at each optimization step. We also discuss our methods for overfitting prevention and navigating production dynamics at scale. These contributions advance the scientific rigor and understanding of LLMs in social applications serving millions of users.

cs.CL↗

arXiv2026

Deep Learning-based Low-Overhead Beam Alignment for mmWave Massive MIMO Systems

Millimeter-wave massive multiple-input multiple-output systems employ highly directional beamforming to overcome severe path loss, and their performance critically depends on accurate beam alignment. Conventional codebook-based methods offer low training overhead but suffer from limited angular resolution and sensitivity to hardware impairments. To address these challenges, we propose a deep learning-enhanced super-resolution beam alignment framework with three key components. First, we design the Quaternary Search-based Super-Resolution (QSSR) algorithm, which leverages the monotonic power ratio property between two discrete Fourier transform (DFT) codebook beams to achieve super-resolution angle estimation without increasing measurement complexity relative to binary search. Second, we develop QSSR-Net, a gated recurrent unit-based neural network that exploits sequential multi-layer beam measurements to capture angular dependencies, thereby improving estimation accuracy, robustness to noise, and generalization across diverse propagation environments. Third, to mitigate the adverse effects of hardware impairments such as antenna position and phase errors, we propose a parametric self-calibration method that requires no additional hardware overhead and adapts compensation parameters in real time. Simulation results show that the proposed framework consistently outperforms binary search and even exhaustive search at high signal-to-noise ratios, achieving substantial performance gains while maintaining low overhead.

eess.SP↗

arXiv2026

Generalizable Coarse-to-Fine Robot Manipulation via Language-Aligned 3D Keypoints

Hierarchical coarse-to-fine policy, where a coarse branch predicts a region of interest to guide a fine-grained action predictor, has demonstrated significant potential in robotic 3D manipulation tasks by especially enhancing sample efficiency and enabling more precise manipulation. However, even augmented with pre-trained models, these hierarchical policies still suffer from generalization issues. To enhance generalization to novel instructions and environment variations, we propose Coarse-to-fine Language-Aligned manipulation Policy (CLAP), a framework that integrates three key components: 1) task decomposition, 2) VLM fine-tuning for 3D keypoint prediction, and 3) 3D-aware representation. Through comprehensive experiments in simulation and on a real robot, we demonstrate its superior generalization capability. Specifically, on GemBench, a benchmark designed for evaluating generalization, our approach achieves a 12\% higher average success rate than the SOTA method while using only 1/5 of the training trajectories. In real-world experiments, our policy, trained on only 10 demonstrations, successfully generalizes to novel instructions and environments.

cs.RO↗