arXiv subjects

Nan Li

Publications and source records attributed to Nan Li.

At least 37 records · Page 2Linked to original sources

arXiv2026

Omni-Photonic Base Station: A Three-Functional-Domain Photonic Architecture for Evolutionary 6G Wireless Infrastructure

6G mobile communications impose immersive communication demands for higher data rates and massive connectivity, while emerging integrated sensing-computing-intelligence scenarios require base stations to concurrently enhance computing capability, guarantee service latency, and realize sensing. We propose the Omni-Photonic Base Station-a progressive evolutionary architecture that introduces photonic technologies into three functional domains, namely baseband processing, fronthaul transmission, and the RF front-end, and obtains system-level gains through cross-domain co-design. By introducing photonics, Omni-PBS harnesses the inherent physical advantages of ultra-broad bandwidth, ultra-low propagation latency, and native parallelism to transcend the aforementioned electronic bottlenecks and fulfill the compound demands of 6G scenarios. The optical baseband computing domain employs photonic accelerators to perform linear computation-intensive tasks, improving the energy efficiency of AI inference by one to two orders of magnitude. The analog optical fronthaul domain replaces digital fronthaul with analog radio-over-fiber , which eliminates the ADCs/DACs and digital intermediate-frequency chips in the remote unit and substantially reduces fronthaul bandwidth requirements. The microwave-photonic RF domain breaks through the bandwidth and frequency limitations of electronic RF front-ends via optical true-time-delay beamforming, programmable photonic filtering, and optical heterodyne frequency conversion. The three domains yield system-level gains beyond single-domain summation through four co-design principles: end-to-end optical-domain continuity, cross-domain co-design, optical computing resource scheduling, and joint optimization of functional splitting.

eess.SP↗

arXiv2026

Personalized Digital Semantic Communication for Image Transmission with Vision-Language Models

Semantic communication (SC) enables bandwidth-efficient wireless image transmission, but most existing SC schemes are user-agnostic and ignore receiver-dependent semantics. To address this issue, we propose a personalized digital semantic communication (PDSC) framework that integrates a vision-language model (VLM)-based semantic encoder with a latent diffusion model (LDM)-based semantic decoder. Specifically, the semantic encoder extracts source-aware personalized semantic tokens from both the source image and the receiver's historical interactions. These tokens are vector-quantized into discrete semantic indices and further encoded into a compact fixed-length bitstream, enabling compatibility with digital transmission. At the receiver, the semantic decoder reconstructs a personalized image conditioned on the recovered semantic tokens. Furthermore, we formulate a capacity-constrained personalized semantic rate-distortion problem and introduce a semantic distortion metric that jointly characterizes source-semantic fidelity and user-preference alignment. Experiments show that PDSC achieves superior source-semantic consistency and personalization over state-of-the-art SC baselines, including CDDM and MoS, under bandwidth-limited wireless transmission.

eess.IV↗

arXiv2026

Lyapunov-Based Completion-Aware Scheduling for PDU Set-Based Real-Time XR Traffic

Real-time extended reality (XR) services impose stringent throughput, latency, and reliability requirements. An XR media unit is commonly segmented into a protocol data unit (PDU) set that is useful only when all required PDUs are delivered within its delay budget. Existing PDU set-aware medium access control (MAC) schedulers generally prioritize transmissions according to urgency or progress, without explicitly accounting for whether a PDU set can still be completed on time. In this paper, we propose a Lyapunov-based completion-aware MAC scheduler that jointly captures the set-level completion dependency, deadline urgency, and completion feasibility of each head-of-line PDU set. Long-term quality-of-service (QoS) requirements are represented through virtual debt queues, and resources are allocated according to their marginal contribution to the timely-completion probability of each head-of-line PDU set. These components are integrated into a drift-plus-penalty scheduling metric that balances long-term reliability requirements against current completion opportunities. System-level simulations demonstrate that the proposed scheduler improves PDU set delivery reliability and lower-tail performance under resource contention, while providing robust performance across heterogeneous traffic demands and network conditions.

eess.SP↗

arXiv2026

Galaxy-LRD Strong Lenses: A Missing Population?

The physical nature of Little Red Dots (LRDs) remains uncertain, although these abundant, compact, and red sources may offer important insights into early black-hole growth and galaxy formation. Strong gravitational lensing can magnify LRDs and spatially resolve their internal structure, thereby helping to discriminate among competing physical scenarios. However, no galaxy-scale strongly lensed LRD has yet been securely confirmed. To predict the abundance of such systems in current and future surveys and to guide dedicated searches, we present the first benchmark estimate of the detectable population of galaxy-scale lensed LRDs by combining literature-based LRD source models with a population of foreground early-type galaxy deflectors. Our Monte Carlo simulation spans $50~{\rm deg}^{2}$ and contains 270,713 LRDs and 5,460,841 deflectors. We predict idealized surface densities of $10.70\pm3.76~{\rm deg}^{-2}$ for doubles and $0.64\pm0.69~{\rm deg}^{-2}$ for quads. After accounting for the JWST point-spread function and survey limiting magnitudes, the detectable surface densities decrease to $3.70\pm1.89~{\rm deg}^{-2}$ and $0.52\pm0.58~{\rm deg}^{-2}$, respectively. For the de-duplicated $0.66~{\rm deg}^{2}$ footprint covered by COSMOS-Web, PRIMER-UDS, PRIMER-COSMOS, CEERS, JADES GOODS-S, and JADES GOODS-N, for which the reported limiting depths are combined through area-weighted averaging in flux space, the predicted probabilities of detecting no systems are $8.6\%$ for doubles and $70.8\%$ for quads.

astro-ph.GA↗

arXiv2026

Twenty-one characterizations of reversible quantum channels

Reversible quantum channels play a fundamental role in quantum dynamics and quantum information processing. A quantum channel is reversible if there exists another quantum channel acting as its left inverse. Due to their intrinsic significance and wide applications, it is desirable to characterize reversible quantum channels from diverse perspectives. In this work, we study reversible quantum channels on finite-dimensional Hilbert spaces, with particular emphasis on the case of different input and output dimensions. We systematically present twenty-one equivalent characterizations of reversible quantum channels from algebraic, geometrical, and information-theoretical perspectives. Among these characterizations, some are well known, while others, implicit in the literature or formulated in other contexts, are clarified here; the Choi-state characterization is derived in this work. Specifically, we prove that the Choi states of reversible quantum channels admit three equivalent forms: the spectral, direct-sum, and tensor-product representations. These twenty-one equivalent characterizations establish a comprehensive framework for reversible quantum channels, provide diverse insights into the structural and information-theoretic properties of quantum channels, and facilitate applications of reversibility in quantum information processing such as quantum error correction, quantum teleportation, and quantum thermodynamics.

quant-ph↗

arXiv2026

CoSimRec: Measuring Coordinated-Content Penetration in Recommender Feedback Loops

Recommender systems shape which content reaches users, making it important to measure whether coordinated activity gains visibility beyond the accounts that initiate it. Existing robustness evaluations largely focus on static target-rank changes and do not capture how coordinated interactions, recommendation, and user response evolve within a feedback loop. We propose CoSimRec, an offline agent-based evaluation framework that models coordinated accounts, dynamic ranking, controlled non-bot responses, and ranking interventions in a shared closed-loop process. CoSimRec introduces the Algorithmic Penetration Rate (APR) metric family: exposure APR is the primary endpoint, while behavior APR is a response-model-conditional sensitivity measure; both can be compared with matched no-attack baselines. We evaluate CoSimRec on MIND, MovieLens, and LastFM with random, popularity-based, feedback-sensitive, MF, BPR-MF, and BPR-LightGCN recommenders. In a risk-blind primary protocol, random controls show no statistically supported positive penetration, whereas popularity-based and feedback-sensitive ranking produce positive APR-Lift in all six master-worker settings, reaching 0.4702 on LastFM. A nine-target MovieLens 1M LightGCN stress test shows positive mean APR-Lift around 25\% injection in all three target-popularity strata, while no-filler profiles remain near zero. Under these controlled conditions, coordinated inputs reach non-bot recommendation slots, providing evidence of a computational pathway from organized activity to audience-level visibility.

cs.IR↗

arXiv2026

Gravitational Lensing Effects by Galaxy Clusters on Ionised Bubble Size Distribution during the Epoch of Reionisation

The statistical properties of ionisation structures during the Epoch of Reionisation (EoR) provide valuable insights into the formation of the first stars and galaxies. However, size distributions of ionisation structures can be affected by gravitational lensing from foreground massive structures such as galaxy clusters. To quantify the impact of cluster lensing on the ionised Bubble Size Distribution (BSD), we performed a series of multiple-lens-plane simulations combining light cones of clusters with source light cones based on different ionisation models. The deflector population is generated with a Monte Carlo method guided by the halo mass function and empirical scaling relations, and the deflectors are modelled with truncated Navarro-Frenk-White (TNFW) profiles. Source light cones are produced semi-numerically or taken directly from the Evolution of 21 cm Structure (EOS) project. Using the Mean Free Path method, we measure both unlensed and lensed BSDs. We find that gravitational lensing increases the apparent abundance of large bubbles while leaving small bubbles nearly unchanged across all source models considered. In particular, for the EOS faint-galaxies model, the apparent number of bubbles with R > 15 cMpc increases by 219% at z = 14; for the EOS bright-galaxies model, it increases by 832% under the same conditions. Moreover, a directional projection test shows only minor lensing-induced changes in line-of-sight direction, suggesting a possible route to recovering unlensed bubble statistics. Above all, lensing introduces unavoidable systematics into BSD measurements that should be carefully taken into account for relevant studies in the SKA era.

astro-ph.CO↗

arXiv2026

Design, Fabrication and Characterization of Microwave Multiplexing SQUID Prototype

The readout system with a high multiplexing ratio has become a bottleneck limiting the application of large-scale Transition Edge Sensor (TES) detector arrays. In recent years, the microwave superconducting quantum interference device (SQUID) multiplexer has emerged as a key technology for effectively reading large-scale cryogenic detector arrays. Currently, the microwave SQUID multiplexer is being adopted by an increasing number of experiments due to its capability of achieving a multiplexing ratio of 2000:1 within the readout bandwidth. In this study, we developed and fabricated a 32-channel microwave SQUID multiplexer prototype. And we measured 8 channels of the prototype. The measured equivalent noise current of the prototype reached 42 pA/$\sqrt{Hz}$.

physics.ins-det↗

arXiv2026

The complexities of patient-centred conversational artificial intelligence

Consumer-facing health chatbots powered by large language models (LLMs) are increasingly used for symptom assessment. However, chatbot development and evaluation often rely on cooperative, articulate, simulated patients. We analysed 2,053 real patient-chatbot conversations and found that communication patterns and expression of emotions vary widely across users. We developed a patient simulator that separately models clinical content, emotional state, conversational strategy, and communication style. In a Turing-inspired evaluation of realism with 15 human graders, simulated conversations were nearly indistinguishable from real ones, with human graders achieving an accuracy of 55%. We used five distinct patient personae, across 1,164 clinician-graded cases, to evaluate the performance of four LLMs in urgency assessment. We found that communication style can significantly alter triage outcomes. Patient-centred conversational artificial intelligence must accommodate communication diversity: systems designed for idealised, rather than realistic, interactions risk underperforming and amplifying health disparities when deployed in the real world.

cs.AI↗

arXiv2026

A Survey of Reinforcement Learning-Based Motion Planning for Autonomous Driving: Lessons Learned from a Driving Task Perspective

Reinforcement learning (RL), with its ability to explore and optimize policies in complex, dynamic decision-making tasks, has emerged as a promising approach to addressing motion planning (MoP) challenges in autonomous driving (AD). Despite rapid advancements in RL and AD, a systematic description and interpretation of the RL design process tailored to diverse driving tasks remains underdeveloped. This survey provides a comprehensive review of RL-based MoP for AD, focusing on lessons from task-specific perspectives. We first outline the fundamentals of RL methodologies, and then survey their applications in MoP, analyzing scenario-specific features and task requirements to shed light on their influence on RL design choices. Building on this analysis, we summarize key design experiences, extract insights from various driving task applications, and provide guidance for future implementations. Additionally, we examine the frontier challenges in RL-based MoP, review recent efforts to addresse these challenges, and propose strategies for overcoming unresolved issues.

cs.LG↗

arXiv2026

Seeing Is Not Sharing: Some Vision-Language Models Overestimate Common Ground in Asymmetric Dialogue

In collaborative dialogue, shared perception does not guarantee shared interpretation. Mutual understanding must be established through interaction. We investigate whether vision-language models (VLMs) can distinguish what could be shared from what has been shared between dialogue participants through grounding. We formulate this as an interpretation-matching task on 13,077 annotated reference expressions from HCRC MapTask dialogues, and evaluate VLMs under systematically controlled manipulations of dialogue context and map-information access. Our results show that providing authentic map images improves overall performance but shifts models toward over-predicting alignment. Textual descriptions of the same map content reproduce this bias, while non-informative images suppress alignment predictions entirely, indicating that the bias is driven by task-relevant map content, not the visual channel. This improvement comes at the cost of degraded accuracy on non-aligned cases. Calibration analysis and reference-chain tracking further suggest that models rely on static referential cues on the maps rather than tracking how grounding unfolds through dialogue history. We observe these patterns most clearly in Qwen3-VL-8B-Instruct and, to varying degrees, in four additional models from two architecture families. In models that exhibit the bias, map content, whether presented visually or textually, is treated as evidence of mutual understanding, conflating potential with established common ground.

cs.CL↗

arXiv2026

Mask-Morph Graph U-Net: A Generalisable Mesh-Based Surrogate for Crashworthiness Field Prediction under Large Geometric Variation

Nonlinear finite element crash simulations are accurate but computationally expensive, limiting their use in iterative design optimisation. Machine-learning surrogate models based on graph neural networks (GNNs) offer a faster alternative. Message-passing GNNs are widely used for mesh simulation, and their shared node and edge update functions are relatively generalisable across varying graph structures. By contrast, non-shareable edge-specific aggregation layers can capture nonlinear relationships more accurately but usually require fixed graph connectivity, which limits generalisability. This paper presents Mask-Morph Graph U-Net (MMGUNet), a practical approach to addressing the limitation of hierarchical Graph U-Net architectures that use edge-specific downsampling and upsampling layers. Fixed coarse graph connectivity is required for edge-specific layers. To retain this while improving spatial correspondence, the proposed method morphs the coarsened graph hierarchy to each input mesh using feature-aligned barycentric parameterisation before constructing cross-graph edges. It further applies node masking during supervised pretraining, followed by parameter-efficient fine-tuning in which high-parameter edge-specific layers are frozen. The proposed approach is evaluated in in-distribution, out-of-distribution, and cross-component transfer settings using mean Euclidean distance and maximum intrusion percentage error. Results show that coarse-graph morphing improves test accuracy relative to a fixed-coarse-graph baseline, while masked supervised pretraining reduces the train-test discrepancy and improves data efficiency during transfer. The proposed model also achieves lower prediction error compared with external baselines. These results demonstrate a practical route toward reusable, data-efficient mesh-based surrogate modelling for crashworthiness design exploration.

cs.LG↗

arXiv2026

A graph neural network surrogate model for mesh-based crashworthiness prediction of vehicle panel components

Crashworthiness is a key performance measure in the design of safety-critical vehicle panel components such as B-pillars. Finite element (FE) simulations are widely used to evaluate crash responses but remain computationally expensive for large-scale, nonlinear impact scenarios, particularly when integrated into iterative design and optimisation processes. Although machine learning-based surrogate models have been developed for rapid crashworthiness analysis, they exhibit limitations in detailed representation of complex 3-dimensional components. Graph Neural Networks (GNNs) have emerged as a promising solution for processing data with complex structures. However, existing GNN models often lack sufficient accuracy and computational efficiency to meet industrial demands. This paper proposes Recurrent Graph U-Net (ReGUNet), a graph-based surrogate model for crashworthiness analysis of vehicle panel components. By representing FE meshes in graph form, the model naturally accommodates complex irregular structural geometries. Its hierarchical architecture improves computational efficiency and accuracy, while the introduction of recurrence enhances stability of temporal predictions over multiple time steps. A side-impact case study of hot-stamped steel B-pillars with varying geometries is used to generate training dataset. The trained model demonstrates high accuracy in predicting the dynamic deformation behaviour and crashworthiness indicators of previously unseen component designs. ReGUNet achieves over a 52% reduction in the average deformation prediction error relative to baseline methods, together with markedly improved computational efficiency. ReGUNet provides rapid and reliable crashworthiness assessments, which in turn accelerates the design cycle of vehicle panel components.

eess.SY↗

arXiv2026

Bounding Curvature Measure on Manifolds with Singularities

Let $X$ be an $n$-dimensional Alexandrov space with curvature $\ge -1$, and let $η> 0$. Define $\mathcal{S}^{k}_η(X)$ as the set of $(k,η)$-singular points in $X$ whose tangent cones are $η$-away from splitting off $\mathbb{R}^{k+1}$ isometrically. For a point $p \in X$, assume that $M = B_2(p) \setminus (\mathcal{S}^{n-2}_η(X) \cup \partial X)$ is a smooth manifold equipped with the Riemannian metric induced by $X$. We prove that the integral of the scalar curvature of $M$ over $B_1(p)$ is bounded from above by a constant depending only on $n$ and $η$. As a special case, this extends Petrunin's bounded curvature integral result for complete manifolds with lower sectional curvature bound to the setting of open manifolds and smooth manifolds with boundary, provided that these manifolds are Alexandrov spaces.

math.DG↗

arXiv2026

TianJi-Environ: An Autonomous AI Scientist for Atmospheric Environmental Research

As atmospheric environmental prediction continues to improve, interpretable validation of pollution mechanisms and feedback processes has become a main challenge in atmospheric chemistry. Yet mechanism validation based on complex numerical models still relies heavily on expert knowledge: mechanistic hypotheses must be operationalized into executable experiments, and model outputs must be organized into traceable evidence. We present TianJi-Environ, an auditable AI Scientist for atmospheric-chemistry mechanism validation. TianJi-Environ establishes the first WRF-Chem-based multi-agent framework that autonomously drives complex atmospheric-chemistry simulations, converting mechanistic hypotheses into executable configurations, testing experiments, and evidence criteria. Using ozone response and particulate-matter feedback as two representative examples, we demonstrate TianJi-Environ's capability for mechanism validation. In a summertime ozone case over the North China Plain, the system detects directionally consistent aerosol-radiation-interaction signals in shortwave radiation and boundary-layer height, but judges the evidence for ozone response to NOx control to be incomplete. In a wintertime PM2.5 case over the Guanzhong Basin, it localizes the unsupported link to insufficient propagation from black-carbon perturbation to particulate response and missing diagnostics of vertical absorptive heating. These results show that TianJi-Environ makes expert-driven mechanism validation explicit, structured, and auditable, offering a reproducible paradigm for multi-agent systems coupled with complex atmospheric-chemistry models.

physics.ao-ph↗

arXiv2026

Thicker amorphous grain boundary complexions reduce plastic strain localization in nanocrystalline Cu-Zr

Amorphous grain boundary complexions have been shown to increase the plasticity of nanocrystalline alloys as compared to ordered grain boundaries. Here, the effect of an important structural descriptor, amorphous complexion thickness, on the plasticity and failure modes of nanocrystalline Cu-Zr is studied with in-situ compression testing, with over 50 micropillars tested. Two model materials were created that differ only in their complexion thickness, with one having a thicker complexion population than the other. The sample with thinner complexions was more likely to experience non-uniform plastic deformation in the form of localized plastic flow or shear banding. In contrast, the sample with thicker complexions displayed more homogeneous plasticity and higher damage tolerance; thicker amorphous complexions suppress localization by absorbing defects. This work demonstrates that increasing complexion thickness can be beneficial for stable plastic flow in nanocrystalline alloys, by improving resistance to strain localization and premature failure.

cond-mat.mtrl-sci↗

arXiv2026

Discovery of a Strong-lens Galaxy Pair with the Smallest Projected Separation

We present the spectroscopic confirmation and lens modeling of HSC~J0233$-$0205, a strong-lensing system produced by a close pair of elliptical galaxies at $z_d=0.790\pm0.022$ that lenses a multi-component background source at $z_s=2.160\pm0.002$. The two deflectors are separated by only $0\arcsec.481\pm0\arcsec.004$ ($3.596\pm0.046$ kpc), making this system a compact galaxy-pair lens at relatively high redshift. Joint five-band HSC lens modeling requires two mass components, with Einstein radii of $0\arcsec.774\pm0\arcsec.011$ and $0\arcsec.767\pm0\arcsec.014$, and yields a circularized Einstein radius of $θ_{\rm E}=1\arcsec.549\pm0\arcsec.018$ for the overall system. The lensed source is reconstructed with three components: two extended components separated by $0\arcsec.462\pm0\arcsec.013$ ($3.830\pm0.108$ kpc), together with a compact component nearly aligned with one of them. Integrating the lensing convergence map within the critical curve gives a projected mass of $M_{\rm lensing,crit}=(9.626\pm0.010)\times10^{11}M_\odot$. Combining this with the stellar mass inferred from SED fitting, $M_{\ast,{\rm crit}}=(1.570\pm0.240)\times10^{11}M_\odot$, we obtain a projected dark-matter fraction within the critical curve of $f_{\rm dm}=83.7\pm2.5\%$. Within the $z$-band effective radii of the two deflectors, the corresponding dark-matter fractions are $82.1\pm4.5\%$ and $65.7\pm6.7\%$, respectively. HSC~J0233$-$0205 is therefore a compact, high-redshift galaxy-pair lens. Future high-resolution imaging and spatially resolved spectroscopy will enable detailed tests of merger signatures, the redistribution of stellar and dark matter, and possible light--mass offsets in the lens plane.

astro-ph.GA↗

arXiv2026

EdgeRazor: A Lightweight Framework for Large Language Models via Mixed-Precision Quantization-Aware Distillation

Quantization has emerged as a mainstream approach for deploying Large Language Models (LLMs) on resource-constrained devices, yet compressing precision below 4-bit typically causes severe performance degradation or prohibitive retraining costs. In this paper, we propose EdgeRazor, a lightweight framework for LLMs via Mixed-Precision Quantization-Aware Distillation. It contains three modules: Structural Quantization with Mixed Precision for fine-grained control of bit-widths, Layer-Adaptive Feature Distillation that dynamically selects the most informative features for alignment, and Entropy-Aware KL Divergence for forward-reverse balance on both human-annotated and distilled datasets. Evaluations conducted on MobileLLM and Qwen families show that under weight-activation quantization, the 1.88-bit Qwen3-0.6B-EdgeRazor outperforms the state-of-the-art 2-bit baselines by 11.27 and surpasses the strongest 3-bit baselines by 4.38, while the quantized MobileLLM-350M-EdgeRazor requires a training budget 4-10$\times$ lower than the leading quantization-aware training method. In terms of efficiency, EdgeRazor achieves higher compression ratios at all bit-widths, and the 1.58-bit Qwen3-0.6B-EdgeRazor reduces storage from 1.11 GB to 0.19 GB while accelerating decoding by 15.16$\times$ over the 16-bit baseline. These results empirically validate the effectiveness and efficiency of EdgeRazor. The codes can be accessed from \href{https://github.com/zhangsq-nju/EdgeRazor}{GitHub} and \href{https://huggingface.co/collections/zhangsq-nju/edgerazor-nbit}{Huggingface}.

cs.LG↗