arXiv subjects

Song Yu

Publications and source records attributed to Song Yu.

At least 19 recordsLinked to original sources

arXiv2026

Discrete-modulated continuous-variable quantum key distribution with uncertainty principle

Continuous-variable quantum key distribution is a compelling framework for scalable quantum networks due to its seamless integration with existing optical communication infrastructure. However, a fundamental gap persists between theoretical protocols requiring ideal Gaussian modulation and the constrained, discrete-modulated signals dictated by practical high-speed hardware. Current security proofs for discrete modulation rely on semidefinite programming, which suffers from prohibitive computational overhead for high-order constellations and lacks direct physical insight into non-Gaussian modulation.In this Letter, we overcome this limitation by developing a security framework that obviates semidefinite programming in favor of an approach grounded fundamentally in the Heisenberg uncertainty principle. By introducing a multi-mode entanglement-source model to characterize non-Gaussian state preparation, we establish an explicit mapping between constellation geometry and the secret key rate. This framework effectively quantifies the security implications of hardware-limited, finite state preparation, enabling both numerical and analytical security analysis under high-order constellations. We experimentally validate our method on both discrete-component and integrated photonic platforms, demonstrating that a quadrature amplitude modulation format with 256 constellation points can asymptotically approach the Gaussian capacity limit. Beyond quantum key distribution, the principle of tightening uncertainty-constrained bounds via source-mode expansion offers a paradigm for exploring the information-theoretic properties of complex non-Gaussian systems.

quant-ph↗

arXiv2026

EP-GRPO: Entropy-Progress Aligned Group Relative Policy Optimization with Implicit Process Guidance

Reinforcement learning with verifiable rewards (RLVR), particularly Group Relative Policy Optimization (GRPO), has advanced LLM reasoning. However, GRPO suffers from three credit assignment failures: uniform token-level granularity that ignores heterogeneous informational value, uniform polarity that penalizes correct steps and rewards incorrect ones, and zero-variance collapse that erases outcome-driven gradients. We systematically quantify these failures, revealing highly non-uniform token informativeness, widespread step-level polarity misalignment, and substantial training waste. To address these limitations, we propose Entropy-Progress Aligned GRPO (EP-GRPO), a framework that mines the model's intrinsic information flow for dense, self-supervised guidance. EP-GRPO integrates entropy-gated modulation to prioritize high entropy decision pivots, implicit process signals from policy divergence anchored to outcome advantages for directional token-level feedback without external reward models, and cumulative entropy mapping that enables progress-aligned advantage normalization, naturally maintaining gradient flow under zero reward variance. Extensive experiments on mathematical reasoning benchmarks demonstrate that EP-GRPO achieves superior accuracy and efficiency compared to GRPO and its variants. The code will be available.

cs.LG↗

arXiv2026

Exploring Boundary-Aware Spatial-Frequency Fusion for Camouflaged Object Detection

Camouflaged Object Detection is challenging due to the high degree of similarity between camouflaged objects and their surrounding backgrounds. Current COD methods mainly rely on edge extraction in the spatial domain and local pixel-level information, neglecting the importance of global structural features. Additionally, they fail to effectively leverage the importance of phase spectrum information within frequency domain features. To this end, we propose a COD framework BASFNet based on boundary-aware frequency domain and spatial domain fusion.This method uses dual guided integration of frequency domain and spatial domain features. A phase-spectrum-based frequency-enhanced edge exploration module (FEEM) and a spatial core segmentation module (SCSM) are introduced to jointly capture the boundary and object features of camouflaged objects. These features are then effectively integrated through a spatial-frequency fusion interaction module (SFFIM). Furthermore, the boundary detection is further optimized through an boundary-aware training strategy. BASFNet outperforms existing state-of-the-art methods on three benchmark datasets, validating the effectiveness of the fusion of frequency and spatial domain information in COD tasks.

cs.CV↗

arXiv2026

ERPO: Token-Level Entropy-Regulated Policy Optimization for Large Reasoning Models

Reinforcement learning from verifiable rewards has significantly advanced the reasoning capabilities of large language models. However, Group Relative Policy Optimization (GRPO) typically assigns a uniform, sequence-level advantage to all tokens, thereby overlooking the intrinsic information heterogeneity along reasoning chains. We show that this coarse-grained credit assignment leads to premature entropy collapse and encourages the model to generate redundant, low-quality reasoning paths. Through systematic empirical analysis, we identify Critical Decision Pivots (CDPs): transient high-entropy states where the policy's trajectory is most sensitive to perturbations. These pivots represent the "forks in the road" where effective multi-path exploration is most crucial yet often suppressed by uniform advantage signals. Building on these insights, we propose Entropy-Regulated Policy Optimization (ERPO), which transitions the optimization focus from coarse sequences to fine-grained token dynamics. ERPO introduces three synergistic components: (i) Entropy-aware Gating, which adaptively amplifies exploration at CDPs to facilitate diverse path discovery; (ii) Bucket-based Implicit Normalization, which mitigates difficulty bias by aligning token progress windows; and (iii) Result-anchored Advantage Synthesis, which re-weights token-level signals via outcome-driven anchors. Extensive experiments on competitive mathematical benchmarks demonstrate that ERPO significantly outperforms GRPO. Notably, ERPO not only boosts reasoning accuracy but also yields significantly more concise and robust derivation paths, while achieving performance comparable to large models with orders of magnitude more parameters.

cs.LG↗

arXiv2026

Practical continuous-variable quantum key distribution using dynamic digital signal processing: security proof and experimental demonstration

Digital signal processing technology has paved the way for the realization of high-speed continuous-variable quantum key distribution systems. However, existing security proofs are limited to static digital signal processing algorithms, while practical systems rely on dynamic multiple-input multiple-output algorithms to compensate for time-varying channel impairments. Our analysis reveals that the conventional dynamic algorithm, due to its non-unitary nature, systematically underestimates the excess noise, which in turn leads to security issues and the generation of insecure keys. To close this gap, we propose a secure algorithm model, mapping the dynamic algorithm to an equivalent physical optical model whose security can be rigorously assessed. Simulations illustrate the algorithm's non-unitary property and provide a quantitative analysis of the excess noise underestimation caused by the conventional algorithm. We further experimentally validate the necessity of the proposed modeling for dynamic digital signal processing, achieving a secret key rate of 14.4 Mbps based on estimated excess noise of 0.07 shot noise unit; whereas the conventional algorithm would have dangerously overestimated the key rate to 28.2 Mbps with noise of 0.008 shot noise unit. This work provides the essential security framework for dynamic digital signal processing, overcoming a critical impediment for the development of high-performance continuous-variable quantum key distribution systems.

quant-ph↗

arXiv2026

Continuous-mode analysis of improved two-way CV-QKD

Continuous-variable quantum key distribution (CV-QKD) enables information-theoretically secure key generation between legitimate parties. To further enhance system performance, an improved two-way CV-QKD protocol has been proposed, which is accessible in practice and exhibits increased robustness against excess noise. However, in practical implementations, device nonidealities inevitably drive the optical field from the single-mode regime into the continuous-mode regime. In this work, we introduce temporal modes to characterize the evolution of optical fields in the improved two-way protocol and establish a security analysis framework for the continuous-mode scenario based on adaptive normalization with calibrated shot-noise unit. In addition, finite-size effects are taken into account in the analysis. Our results demonstrate that the improved two-way protocol retains a performance advantage over one-way counterpart. The analysis provides useful guidance for the practical implementation and performance optimization of improved two-way CV-QKD systems.

quant-ph↗

arXiv2025

Analyzing the performance of CV-MDI QKD under continuous-mode scenarios

Continuous-variable measurement-device-independent quantum key distribution (CV-MDI QKD) can address vulnerabilities on the detection side of a QKD system. The core of this protocol involves continuous-variable Bell measurements performed by an untrusted third party. However, in high-speed systems, spectrum broadening causes Bell measurements to deviate from the ideal single-mode scenario, resulting in mode mismatches, reduced performance, and compromised security. Here, we introduce temporal modes (TMs) to analyze the performance of CV-MDI QKD under continuous-mode scenarios. The mismatch between Bob's transmitting mode and Bell measurement mode has a more significant effect on system performance compared to that on Alice's side. When the Bell receiver is close to Bob and the mismatch is set to just 5%, the transmission distance drastically decreases from 87.96 km to 18.50 km. In comparison, the same mismatch for Alice reduces the distance to 86.83 km. This greater degradation on Bob's side can be attributed to the asymmetry in the data modification step. Furthermore, the mismatch in TM characteristics leads to a significant reduction in the secret key rate by 83% when the transmission distance is set to 15 km, which severely limits the practical usability of the protocol over specific distances. These results indicate that in scenarios involving continuous-mode interference, such as large-scale MDI network setups, careful consideration of each user's TM characteristics is crucial. Rigorous pre-calibration of these modes is essential to ensure the system's reliability and efficiency.

quant-ph↗

arXiv2025

Multi-Component Open/Relative/Local Correspondence

For a toric Calabi-Yau 3-orbifold relative to s Aganagic-Vafa outer branes, we prove a correspondence among the genus-zero open Gromov-Witten invariants with maximal winding at each brane and: (i) closed invariants of a toric Calabi-Yau (3+s)-orbifold; (ii) formal relative invariants of a formal toric Calabi-Yau (FTCY) 3-orbifold with maximal tangency to s divisors; (iii) formal relative invariants of a sequence of FTCY intermediate geometries interpolating dimensions 3 and 3+s. The correspondence provides examples of the log/local principle of van Garrel-Graber-Ruddat in the multi-component setting and the refined conjecture of Brini-Bousseau-van Garrel via intermediate geometries. It also establishes the multi-component case of the open/closed correspondence proposed by Lerche-Mayr and studied by Liu-Yu. As an application, we obtain examples of the conjecture of Klemm-Pandharipande on the integrality of BPS invariants of higher-dimensional toric Calabi-Yau manifolds. Along the way, we set the basic stages of the relative Gromov-Witten theory of higher-dimensional FTCY orbifolds, generalizing the case of smooth 3-folds by Li-Liu-Liu-Zhou.

math.AG↗

arXiv2025

Continuous-mode analysis for practical continuous-variable quantum key distribution

Continuous-variable quantum key distribution (CV-QKD) enables two remote parties to establish information-theoretically secure keys and offers high practical feasibility due to its compatibility with mature coherent optical communication technologies. However, as CV-QKD systems progress toward digital implementations, device nonidealities drive the optical field from a single-mode to a continuous-mode region, thereby underscoring the mismatch between theoretical models and practical systems. Here, we introduce temporal modes to construct an entanglement-based scheme that more accurately captures device nonidealities and develop a corresponding secret key rate calculation method applicable to continuous-mode scenarios. We demonstrate that optimizing the pulse-shaping format can significantly improve performance under detector-bandwidth-limited conditions. Experimental results also confirm that the proposed model effectively describes the impact of sampling-time deviations. We further analyze a linear weighted-reconstruction digital signal processing method,which improves the secret key rate by approximately 50% in a 30-km fiber experiment without requiring additional hardware, demonstrating a substantial performance enhancement at metropolitan distances. The proposed theoretical framework accommodates a broader range of experimental conditions and can guide the optimization of digital CV-QKD systems.

quant-ph↗

arXiv2025

DGAI: Decoupled On-Disk Graph-Based ANN Index for Efficient Updates and Queries

On-disk graph-based indexes are favored for billion-scale Approximate Nearest Neighbor Search (ANNS) due to their high performance and cost-efficiency. However, existing systems typically rely on a coupled storage architecture that co-locates vectors and graph topology, which introduces substantial redundant I/O during index updates, thereby degrading usability in dynamic workloads. In this paper, we propose a decoupled storage architecture that physically separates heavy vectors from the lightweight graph topology. This design substantially improves update performance by reducing redundant I/O during updates. However, it introduces I/O amplification during ANNS, leading to degraded query efficiency.To improve query performance within the update-friendly architecture, we propose two techniques co-designed with the decoupled storage. We develop a similarity-aware dynamic layout that optimizes data placement online so that redundantly fetched data can be reused in subsequent search steps, effectively turning read amplification into useful prefetching. In addition, we propose a two-stage query mechanism enhanced by hierarchical PQ, which uses hierarchical PQ to rapidly and accurately identify promising candidates and performs exact refinement on raw vectors for only a small number of candidates. This design significantly reduces both the I/O and computational cost of the refinement stage. Overall, DGAI achieves resource-efficient updates and low-latency queries simultaneously. Experimental results demonstrate that \oursys improves update speed by 8.17x for insertions and 8.16x for deletions, while reducing peak query latency under mixed workloads by 67\% compared to state-of-the-art baselines.

cs.DB↗

arXiv2025

Tree of Agents: Improving Long-Context Capabilities of Large Language Models through Multi-Perspective Reasoning

Large language models (LLMs) face persistent challenges when handling long-context tasks, most notably the lost in the middle issue, where information located in the middle of a long input tends to be underutilized. Some existing methods that reduce input have the risk of discarding key information, while others that extend context windows often lead to attention dispersion. To address these limitations, we propose Tree of Agents (TOA), a multi-agent reasoning framework that segments the input into chunks processed by independent agents. Each agent generates its local cognition, then agents dynamically exchange information for collaborative reasoning along tree-structured paths. TOA enables agents to probe different reasoning orders for multi-perspective understanding, effectively mitigating position bias and reducing hallucinations. To improve processing efficiency, we incorporate prefix-hash caching and adaptive pruning strategies, achieving significant performance improvements with comparable API overhead. Experiments show that TOA, powered by compact LLaMA3.1-8B, significantly outperforms multiple baselines and demonstrates comparable performance to the latest and much larger commercial models, such as Gemini1.5-pro, on various long-context tasks. Code is available at https://github.com/Aireduce952/Tree-of-Agents.

cs.AI↗

arXiv2025

GeoLayer: Towards Low-Latency and Cost-Efficient Geo-Distributed Graph Stores with Layered Graph

The inherent connectivity and dependency of graph-structured data, combined with its unique topology-driven access patterns, pose fundamental challenges to conventional data replication and request routing strategies in geo-distributed cloud storage systems. In this paper, we propose GeoLayer, a geo-distributed graph storage framework that jointly optimizes graph replica placement and pattern request routing. We first construct a latency-aware layered graph architecture that decomposes the graph topology into multiple layers, aiming to reduce the decision space and computational complexity of the optimization problem, while mitigating the impact of network heterogeneity in geo-distributed environments. Building on the layered graph, we introduce an overlap-centric replica placement scheme to accommodate the diversity of graph pattern accesses, along with a directed heat diffusion model that captures heat conduction and superposition effects to guide data allocation. For request routing, we develop a stepwise layered routing strategy that performs progressive expansion over the layered graph to efficiently retrieve the required data. Experimental results show that, compared to state-of-the-art replica placement and routing schemes, GeoLayer achieves a 1.34x - 3.67x improvement in response times for online graph pattern requests and a 1.28x - 3.56x speedup in offline graph analysis performance.

cs.DB↗

arXiv2025

GoVector: An I/O-Efficient Caching Strategy for High-Dimensional Vector Nearest Neighbor Search

Graph-based high-dimensional vector indices have become a mainstream solution for large-scale approximate nearest neighbor search (ANNS). However, their substantial memory footprint often requires storage on secondary devices, where frequent on-demand loading of graph and vector data leads to I/O becoming the dominant bottleneck, accounting for over 90\% of query latency. Existing static caching strategies mitigate this issue only in the initial navigation phase by preloading entry points and multi-hop neighbors, but they fail in the second phase where query-dependent nodes must be dynamically accessed to achieve high recall. We propose GoVector, an I/O-efficient caching strategy tailored for disk-based graph indices. GoVector combines (1) a static cache that stores entry points and frequently accessed neighbors, and (2) a dynamic cache that adaptively captures nodes with high spatial locality during the second search phase. To further align storage layout with similarity-driven search patterns, GoVector reorders nodes on disk so that similar vectors are colocated on the same or adjacent pages, thereby improving locality and reducing I/O overhead. Extensive experiments on multiple public datasets show that GoVector achieves substantial performance improvements. At 90% recall, it reduces I/O operations by 46% on average, increases query throughput by 1.73x, and lowers query latency by 42% compared to state-of-the-art disk-based graph indexing systems.

cs.DB↗

arXiv2025

Quantum cohomology of variations of GIT quotients and flips

We prove a decomposition theorem for the quantum cohomology of variations of GIT quotients. More precisely, for any reductive group $G$ and a simple $G$-VGIT wall-crossing $X_- \dashrightarrow X_+$ with a wall $S$, we show that the quantum $D$-module of $X_-$ can be decomposed into a direct sum of that of $X_+$ and copies of that of $S$. As an application, we obtain a decomposition theorem for the quantum cohomology of local models of standard flips in birational geometry.

math.AG↗

arXiv2025

Hodge-theoretic Open/Closed Correspondence

We continue the B-model development of the open/closed correspondence proposed by Mayr and Lerche-Mayr, complementing the A-model study in the preceding joint works with Liu and providing a Hodge-theoretic perspective. Given a corresponding pair of open geometry on a toric Calabi-Yau 3-orbifold $\mathcal{X}$ relative to a framed Aganagic-Vafa brane $\mathcal{L}$ and closed geometry on a toric Calabi-Yau 4-orbifold $\widetilde{\mathcal{X}}$, we consider the Hori-Vafa mirrors $\mathcal{X}^\vee$ and $\widetilde{\mathcal{X}}^\vee$, where the mirror of $\mathcal{L}$ can be given by a family of hypersurfaces $\mathcal{Y} \subset \mathcal{X}^\vee$. We show that the Picard-Fuchs system associated to $\widetilde{\mathcal{X}}$ extends that associated to $\mathcal{X}$ and characterize the full solution space in terms of the open string data. Furthermore, we construct a correspondence between integral 4-cycles in $\widetilde{\mathcal{X}}^\vee$ and relative 3-cycles in $(\mathcal{X}^\vee, \mathcal{Y})$ under which the periods of the former match the relative periods of the latter. On the dual side, we identify the variations of mixed Hodge structures on the middle-dimensional cohomology of $\widetilde{\mathcal{X}}^\vee$ with that on the middle-dimensional relative cohomology of $(\mathcal{X}^\vee, \mathcal{Y})$ up to a Tate twist.

math.AG↗

arXiv2025

Approaching the Key Rate Limit in Continuous-Variable Quantum Key Distribution Network

A quantum key distribution network enables pairs of users to generate independent secret keys by leveraging the principles of quantum physics. For end-to-end secure communication, a user pair's secret key must remain secure against any third parties, including both external eavesdroppers and other network users. However, isolating a given user pair from the remaining users while maintaining a high key rate is challenging when all users are intrinsically coupled and correlated, particularly in continuous-variable networks. This results in either a low key rate, or incomplete end-to-end security. Here, we introduce a multi-user security framework, offering a general and comprehensive end-to-end key rate formula, against the collaboration of the other network users and eavesdropper. Building on this framework, we propose a multi-user protocol that achieves the theoretical upper limit in all practical deployments within a 100 km range. Applied to a three-node network, it achieves an Mbps-level per-user key rate and an overall network key rate reaching 90\% of the upper limit. The proposed solution supports scalable implementations using telecom-compatible components, while the method for obtaining the accessible information in the network is broadly applicable and can be extended to various multipartite quantum information systems.

quant-ph↗

arXiv2025

Fully Integrated Vacuum-based Quantum Random Number Generator

Quantum random number generators play a crucial role in securing high-demand information contexts by producing true random numbers. Nevertheless, the large volume and high-cost limit their widespread use. Here, we propose a system on chip that fully leverages the advantages of different photonic integrated platforms, where the interference optical paths and photodiodes are integrated on a standard silicon process, while the laser source on-chip is realized on a III-V platform. Using micro-lens coupling package technology, which contributes to a topnotch coupling loss lower than 2dB, the components on different platforms are combined and packaged with the amplifier circuits in a 42mm* 24mm footprint in a butterfly form. This complete miniaturized and cost-effective entropy source enables outputting a vacuum noise signal with a 3dB bandwidth of over 500MHz. After sampling and post-processing, a random number generation rate of up to 6.57Gbps is achieved. The results show a feasible way of overcoming the laser integration problem with silicon-based integrated quantum photonics. Foreseeable, commercial applications on a large scale are significantly promoted.

physics.optics↗

arXiv2025

Remodeling Conjecture with Descendants

We formulate and prove the Remodeling Conjecture with descendants, which is a version of all-genus equivariant descendant mirror symmetry for semi-projective toric Calabi-Yau 3-orbifolds with integral structures. We construct an isomorphism between the $K$-group of equivariant coherent sheaves on the toric Calabi-Yau 3-orbifold with support bounded in a direction and a certain integral relative first homology group of the equivariant mirror curve. Under this isomorphism, we prove the equivariant mirror symmetric Gamma conjecture which equates quantum cohomology central charges of coherent sheaves and oscillatory integrals along corresponding relative 1-cycles. As a consequence in the non-equivariant setting, we prove a conjecture of Hosono which equates central charges of compactly supported coherent sheaves and period integrals of integral 3-cycles on the Hori-Vafa mirror 3-fold. Furthermore, we establish a correspondence between all-genus equivariant descendant Gromov-Witten invariants with $K$-theoretic framings and oscillatory integrals (Laplace transforms) of the Chekhov-Eynard-Orantin topological recursion invariants along relative 1-cycles on the equivariant mirror curve.

math.AG↗