arXiv subjects

Si Shen

Publications and source records attributed to Si Shen.

At least 19 recordsLinked to original sources

arXiv2026

Entanglement-based quantum key distribution with data in hollow-core fiber

The coexistence of quantum information and classical signals in a single fiber is essential for future quantum networks that leverage the well-established optical fiber infrastructure. Although multiplexing technologies can separate quantum and classical signals, pure silica core fibers (PSCFs) remain fundamentally limited by the high nonlinearity, which generates substantial Raman scattering and four-wave mixing noise. Hollow-core fibers (HCFs), guiding light predominantly in air, offer an attractive solution with intrinsically ultra-low nonlinearity and strongly suppressed nonlinear noise. In this work, we demonstrate the entanglement-based key coexisting with data over an 18-km HCF link. We achieve time-encoded high-dimensional quantum key distribution (HD-QKD) carrying 0 dBm of bidirectional received power, corresponding to a theoretical data capacity of up to 2.3 Tbps. During 24 hours of continuous operation, an average secret key rate (SKR) of 10.56 kbps is obtained. Theoretical analysis further predicts SKRs above 135 kbps over transmission distances exceeding 200 km using state-of-the-art low-loss HCFs. These results show significantly improved performance compared with PSCF-based systems and highlight the potential of HCFs for scalable quantum-classical coexistence compatible with the architectures of established fiber-optic networks.

quant-ph↗

arXiv2026

Quantum teleportation over a field-deployed hollow-core fibre network

When a photon and one member of an entangled photon pair are jointly projected onto a Bell-state measurement (BSM), the quantum state of the photon can be transferred to the distant partner of the pair without physically transmitting this information carrier. In real-world deployment, however, teleportation performance is fundamentally bottlenecked by quantum channel impairments, such as loss, noise, and fluctuations, which induce severe decoherence and degrade fidelity. This vulnerability is further exacerbated in scenarios with intense classical data traffic or background light. Realizing scalable quantum networks, therefore, hinges on developing advanced channel architectures capable of supporting both high-fidelity quantum operations and high-capacity classical communications within a shared infrastructure. Towards this end, hollow core fibre (HCF) offers a promising quantum channel resource by combining free-space-like weak light-matter interaction with the stability of fibre-based systems. Here, utilizing a field-deployed metropolitan HCF network spanning three spatially separated nodes in Chengdu, we achieve quantum teleportation with an intermediate BSM under co-propagating classical traffic. Crucially, the HCF links preserve the long-term indistinguishability of photonic qubits without active stabilization, and exhibit a Raman noise approximately three orders of magnitude lower than that of standard solid-core counterparts. This noise suppression enables robust quantum teleportation even alongside classical launch powers up to 160 mW. Our findings establish a classical-data-compatible framework for quantum networking over deployed fibre infrastructure and offer a wavelength-agnostic, plug-and-play, and free-running pathway toward the quantum internet.

quant-ph↗

arXiv2026

Quantum Teleportation toward the Quantum Internet: A Concise Review

Quantum networks play a pivotal role in quantum information science, which not only provide a secure communication platform for remote access to quantum computers but also serve as the strategic core for achieving large-scale quantum information processing, forming the foundational infrastructure for the future global-scale quantum internet. Quantum teleportation, which enables the transmission of unknown quantum states over long distances by employing quantum entanglement together with classical communication, is essential for the distribution of quantum resources in the construction of the global-scale quantum internet. To realize a global-scale quantum internet, quantum repeater protocols represent one of the most promising approaches for enabling quantum communication between any nodes. This concise review presents representative experimental demonstrations of quantum teleportation for constructing quantum networks across different physical platforms. Along this trajectory, the review discusses current challenges, open issues, and future perspectives toward scalable and practical quantum internet.

quant-ph↗

arXiv2026

Verify, Repair, Repeat, or Stop? Robust Stopping for Noisy Verify-Repair Loops in LLM Agents

Verify-repair loops are a standard means for large language model (LLM) agents to correct faulty plans in code generation, mathematical reasoning, and tool use. When both the verifier and the repairer are noisy, repair can damage already-correct plans, and reported acceptance keeps rising while true validity falls, so existing methods lack a principled basis for deciding when repair should stop. We propose VRR-Stop, a robust stopping framework for noisy verify-repair-repeat (VRR) loops. A four-parameter noise model separates verifier false acceptance and false rejection from the repair and damage behavior of the repairer. Belief filtering turns repeated verification votes into an estimate of committed validity, and the loop commits or repairs according to the sign of the true marginal gain, which requires only sign identifiability rather than accurate recovery of all parameters. When verifier discrimination approaches zero, calibration itself fails and estimation error can flip the stopping sign, so we pair VRR-Stop with VRR-Guard, an estimation-free fallback that replaces the incumbent candidate only under a sufficient verification margin. On a GSM8K stress setting, VRR-Stop improves final true validity by 60.6 percentage points over fixed five-round repair at an average cost of 0.72 repair rounds. Across settings, stopping reliability is governed jointly by verifier discrimination and the decision margin rather than by the absolute size of estimation error.

cs.AI↗

arXiv2026

UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods

The deployment of Mixture-of-Experts (MoE) models on production high-bandwidth superpods, such as NVIDIA's NVL72/576 and Huawei's CloudMatrix384, introduces critical challenges beyond raw interconnect bandwidth. While these systems provide unified global address spaces and high-bandwidth fabrics, their full potential for sparse MoE communication is hindered by three fundamental bottlenecks: (1) Strict execution serialization imposed by coarse-grained Bulk Synchronous Parallel (BSP) orchestration of interdependent communication phases; (2) Prohibitive synchronization overhead that fails to scale alongside high interconnect bandwidth; and (3) Severe load imbalance resulting from distance-agnostic scheduling of irregular token traffic. To eliminate these bottlenecks, we introduce UBEP (Unified-Bus Expert Parallelism), a production-ready communication library that rethinks MoE's All-to-All primitives for modern superpod architectures. Through large scale experiments, UBEP reduces All-to-All latency by up to 52.4% and MoE inference Time Per Output Token (TPOT) by up to 11.1%.

cs.DC↗

arXiv2026

Quantum LiDAR with non-local modulation

Quantum light detection and ranging (LiDAR) utilizes quantum entanglement and correlation to improve precision, noise resilience and covertness of target detection. Despite recent advances, the development of a quantum LiDAR system that simultaneously achieves high precision and a large measurement range remains challenging. Here, we demonstrate a quantum amplitude-modulated continuous wave LiDAR with micrometer precision achievable via increased acquisition time and meter-scale measurement range. In our demonstration, the signal photons directly illuminate the target, while the idler photons are non-locally modulated with a high-frequency cosine wave and never interact with the target. By leveraging the non-local modulation and the quantum correlation, the target detection is achieved with a precision of 0.64 $\pm$ 0.06 mm within one second over a measurement range of 2-8 m. As the acquisition time is up to 500 s, the system achieves a precision of 29 $\pm\ 4{\ \mathrm{\mu m}}$. Furthermore, our system realizes a 50 times precision improvement over the classical single-photon scheme in a background noise 37 dB stronger than the returned probe photons. With these advantages, our method will open venues for the development of high-precision, long-range, and noise-resilient target detection.

quant-ph↗

arXiv2026

LLM-Metrics: Measuring Research Impact Through Large Language Model Memory

Citation counts remain the dominant metric for assessing research impact, yet they suffer from well-documented limitations: temporal lag, disciplinary bias, and Matthew effects. Here we propose LLM-Metrics, a research-impact assessment metric derived from the parametric memory of large language models (LLMs). The central hypothesis is that high-impact papers receive greater exposure in the academic community, that this exposure enters LLM training data in textual form, and that models consequently form stronger parametric memory of these papers. We designed four types of multiple-choice probes, covering title recognition, author recognition, method recognition, and venue recognition, and evaluated 549 computer science papers published in 2023-2024 across 17 LLMs spanning 0.5B to 72B parameters from six vendors. Of the 17 models, 15 produced positive predictions, 9 of which were significant at p less than 0.05, with an overall Spearman correlation of rho = 0.1495 and p = 0.0004 against citation counts. Three additional findings support the proposed mechanism. First, the predictive signal was stronger for 2024 papers, rho = 0.1880, whose citation counts were near zero at model-training time, reducing the plausibility of a simple reverse-causality explanation. Second, author-recognition probes showed the strongest discriminative power, consistent with an exposure-driven memory mechanism. Third, model scale and predictive power were non-monotonic: a 3B-parameter model, Llama-3.2-3B-Instruct, with rho = 0.1829, outperformed most larger models, supporting a selective-memory hypothesis in which the limited capacity of smaller models can serve as an effective information filter. LLM-Metrics offers a real-time, cross-disciplinary, citation-independent paradigm for research assessment.

cs.AI↗

arXiv2026

VersaVogue: Visual Expert Orchestration and Preference Alignment for Unified Fashion Synthesis

Diffusion models have driven remarkable advancements in fashion image generation, yet prior works usually treat garment generation and virtual dressing as separate problems, limiting their flexibility in real-world fashion workflows. Moreover, fashion image synthesis under multi-source heterogeneous conditions remains challenging, as existing methods typically rely on simple feature concatenation or static layer-wise injection, which often causes attribute entanglement and semantic interference. To address these issues, we propose VersaVogue, a unified framework for multi-condition controllable fashion synthesis that jointly supports garment generation and virtual dressing, corresponding to the design and showcase stages of the fashion lifecycle. Specifically, we introduce a trait-routing attention (TA) module that leverages a mixture-of-experts mechanism to dynamically route condition features to the most compatible experts and generative layers, enabling disentangled injection of visual attributes such as texture, shape, and color. To further improve realism and controllability, we develop an automated multi-perspective preference optimization (MPO) pipeline that constructs preference data without human annotation or task-specific reward models. By combining evaluators of content fidelity, textual alignment, and perceptual quality, MPO identifies reliable preference pairs, which are then used to optimize the model via direct preference optimization (DPO). Extensive experiments on both garment generation and virtual dressing benchmarks demonstrate that VersaVogue consistently outperforms existing methods in visual fidelity, semantic consistency, and fine-grained controllability.

cs.CV↗

arXiv2025

Jointly Conditioned Diffusion Model for Multi-View Pose-Guided Person Image Synthesis

Pose-guided human image generation is limited by incomplete textures from single reference views and the absence of explicit cross-view interaction. We present jointly conditioned diffusion model (JCDM), a jointly conditioned diffusion framework that exploits multi-view priors. The appearance prior module (APM) infers a holistic identity preserving prior from incomplete references, and the joint conditional injection (JCI) mechanism fuses multi-view cues and injects shared conditioning into the denoising backbone to align identity, color, and texture across poses. JCDM supports a variable number of reference views and integrates with standard diffusion backbones with minimal and targeted architectural modifications. Experiments demonstrate state of the art fidelity and cross-view consistency.

cs.CV↗

arXiv2025

Direct Observation and Optical Manipulation of Exciton-polariton Parametric Scattering Lasing in Temporal

The hybrid light-matter character of exciton-polaritons gives rise to distinct polariton parametric scattering (PPS) process, which holds promise for frontier applications in polaritonic quantum devices. However, the stable excitation and coherent optical manipulation of PPS remain challenging due to scattering bottlenecks and rapid dephasing effect in polariton many-body systems. In this study, we first report the direct observation and optical amplification of non-degenerate intermode PPS lasing at room temperature (RT). The specific polariton branch of strong-coupled nanobelt planar microcavity is resonantly excited by a near-infrared (NIR) femtosecond laser via two-photon absorption (TPA) scheme, and the non-degenerate signal- and idler-states are stimulated. Angle-resolved dispersion patterns clearly reveal the evolution of the pump-, signal-, and idler-states under different excitation powers. Based on our self-constructed ultrafast femtosecond resonant optical trigger set-up, a selective enhancement and modulation of the signal-state is realized. Furthermore, the dynamic measurements of nonlinear signal-state enhancement process demonstrate a sub-picosecond response time (0.4ps), confirming its potential for ultrafast optical manipulation. Our work establishes a platform for exploring TPA-driven PPS laser and provides a novel optical modulation route for polariton-based optoelectronic devices.

physics.optics↗

arXiv2025

Mitigating Think-Answer Mismatch in LLM Reasoning Through Noise-Aware Advantage Reweighting

Group-Relative Policy Optimization (GRPO) is a key technique for training large reasoning models, yet it suffers from a critical vulnerability: the \emph{Think-Answer Mismatch}, where noisy reward signals corrupt the learning process. This problem is most severe in unbalanced response groups, paradoxically degrading the signal precisely when it should be most informative. To address this challenge, we propose Stable Group-Relative Policy Optimization (S-GRPO), a principled enhancement that derives optimal, noise-aware advantage weights to stabilize training. Our comprehensive experiments on mathematical reasoning benchmarks demonstrate S-GRPO's effectiveness and robustness. On various models, S-GRPO significantly outperforms DR. GRPO, achieving performance gains of +2.5% on Qwen-Math-7B-Base, +2.2% on Llama-3.2-3B-Base, and +2.4% on Qwen-Math-1.5B-Instruct. Most critically, while standard GRPO fails to learn under 20% synthetic reward noise, S-GRPO maintains stable learning progress. These results highlight S-GRPO's potential for more robust and effective training of large-scale reasoning models. \footnote{Code and data are available at: https://github.com/shenpeijun0212/S-GRPO

cs.LG↗

arXiv2025

Long Is More Important Than Difficult for Training Reasoning Models

Difficult problems, which often result in long reasoning traces, are widely recognized as key factors for enhancing the performance of reasoning models. However, such high-challenge problems are scarce, limiting the size of available datasets. In this paper, we propose a simple method to decouple the reliance on problem difficulty. First, we empirically demonstrate that reasoning length, rather than problem difficulty, primarily influences the performance of trained models. Second, we identify a scaling law on reasoning length, showing that model performance increases in a log-linear fashion as the reasoning data length grows. Finally, we introduce a straightforward technique to generate reasoning data of arbitrary length, and show that synthesized data is effective for training reasoning models. After fine-tuning the Qwen2.5-32B-Instruct language model on our Long1K dataset, we present our model, Long1K-32B, which achieves remarkable performance with only 1,000 training samples, achieving 95.6\% accuracy on MATH, and 71.1\% on GPQA outperforming DeepSeek-R1-Distill-Qwen-32B. The model, code, and dataset are all open-sourced, available at https://huggingface.co/ZTss/LONG1.

cs.CL↗

arXiv2024

RevOrder: A Novel Method for Enhanced Arithmetic in Language Models

This paper presents RevOrder, a novel technique aimed at improving arithmetic operations in large language models (LLMs) by reversing the output digits in addition, subtraction, and n-digit by 1-digit (nD by 1D) multiplication tasks. Our method significantly reduces the Count of Sequential Intermediate Digits (CSID) to $\mathcal{O}(1)$, a new metric we introduce to assess equation complexity. Through comprehensive testing, RevOrder not only achieves perfect accuracy in basic arithmetic operations but also substantially boosts LLM performance in division tasks, particularly with large numbers where traditional models struggle. Implementation of RevOrder is cost-effective for both training and inference phases. Moreover, applying RevOrder to fine-tune the LLaMA2-7B model on the GSM8K math task results in a considerable improvement, reducing equation calculation errors by 46% and increasing overall scores from 41.6 to 44.4.

cs.AI↗

arXiv2023

GujiBERT and GujiGPT: Construction of Intelligent Information Processing Foundation Language Models for Ancient Texts

In the context of the rapid development of large language models, we have meticulously trained and introduced the GujiBERT and GujiGPT language models, which are foundational models specifically designed for intelligent information processing of ancient texts. These models have been trained on an extensive dataset that encompasses both simplified and traditional Chinese characters, allowing them to effectively handle various natural language processing tasks related to ancient books, including but not limited to automatic sentence segmentation, punctuation, word segmentation, part-of-speech tagging, entity recognition, and automatic translation. Notably, these models have exhibited exceptional performance across a range of validation tasks using publicly available datasets. Our research findings highlight the efficacy of employing self-supervised methods to further train the models using classical text corpora, thus enhancing their capability to tackle downstream tasks. Moreover, it is worth emphasizing that the choice of font, the scale of the corpus, and the initial model selection all exert significant influence over the ultimate experimental outcomes. To cater to the diverse text processing preferences of researchers in digital humanities and linguistics, we have developed three distinct categories comprising a total of nine model variations. We believe that by sharing these foundational language models specialized in the domain of ancient texts, we can facilitate the intelligent processing and scholarly exploration of ancient literary works and, consequently, contribute to the global dissemination of China's rich and esteemed traditional culture in this new era.

cs.CL↗

arXiv2023

Discrete frequency-bin entanglement generation via cascaded second-order nonlinear processes in Sagnac interferometer

Discrete frequency-bin entanglement is an essential resource for applications in quantum information processing. In this Letter, we propose and demonstrate a scheme to generate discrete frequency-bin entanglement with a single piece of periodically poled lithium niobate waveguide in a modified Sagnac interferometer. Correlated two-photon states in both directions of the Sagnac interferometer are generated through cascaded second-order optical nonlinear processes. A relative phase difference between the two states is introduced by changing the polarization state of pump light, thus manipulating the two-photon state at the output of the Sagnac interferometer. The generated two-photon state is sent into a fiber polarization splitter, then a pure discrete frequency-bin entangled two-photon state is obtained by setting the pump light. The frequency entanglement property is measured by a spatial quantum beating with a visibility of $96.0 \pm 6.1\%$. The density matrix is further obtained with a fidelity of $98.0 \pm 3.0\%$ to the ideal state. Our demonstration provides a promising method for the generation of pure discrete frequency-bin entanglement at telecom band, which is desired in quantum photonics.

quant-ph↗

arXiv2023

Hertz-rate metropolitan quantum teleportation

Quantum teleportation can transfer an unknown quantum state between distant quantum nodes, which holds great promise in enabling large-scale quantum networks. To advance the full potential of quantum teleportation, quantum states must be faithfully transferred at a high rate over long distance. Despite recent impressive advances, a high-rate quantum teleportation system across metropolitan fiber networks is extremely desired. Here, we demonstrate a quantum teleportation system which transfers quantum states carried by independent photons at a rate of 7.1$\pm$0.4 Hz over 64-km-long fiber channel. An average single-photon fidelity of $\geqslant$ 90.6$\pm$2.6% is achieved, which exceeds the maximum fidelity of 2/3 in classical regime. Our result marks an important milestone towards quantum networks and opens the door to exploring quantum entanglement based informatic applications for the future quantum internet.

quant-ph↗

arXiv2022

Increasing Visual Awareness in Multimodal Neural Machine Translation from an Information Theoretic Perspective

Multimodal machine translation (MMT) aims to improve translation quality by equipping the source sentence with its corresponding image. Despite the promising performance, MMT models still suffer the problem of input degradation: models focus more on textual information while visual information is generally overlooked. In this paper, we endeavor to improve MMT performance by increasing visual awareness from an information theoretic perspective. In detail, we decompose the informative visual signals into two parts: source-specific information and target-specific information. We use mutual information to quantify them and propose two methods for objective optimization to better leverage visual signals. Experiments on two datasets demonstrate that our approach can effectively enhance the visual awareness of MMT model and achieve superior results against strong baselines.

cs.CV↗

arXiv2022

SsciBERT: A Pre-trained Language Model for Social Science Texts

The academic literature of social sciences records human civilization and studies human social problems. With its large-scale growth, the ways to quickly find existing research on relevant issues have become an urgent demand for researchers. Previous studies, such as SciBERT, have shown that pre-training using domain-specific texts can improve the performance of natural language processing tasks. However, the pre-trained language model for social sciences is not available so far. In light of this, the present research proposes a pre-trained model based on the abstracts published in the Social Science Citation Index (SSCI) journals. The models, which are available on GitHub (https://github.com/S-T-Full-Text-Knowledge-Mining/SSCI-BERT), show excellent performance on discipline classification, abstract structure-function recognition, and named entity recognition tasks with the social sciences literature.

cs.CL↗